Skip to main content

Full text of "A Critical Analysis of the Sutta Nipata"

See other formats

A Critical Analysis of the Sutta Nipata - N.A. Jayawikrama 

A series of articles from the Pali Buddhist Review 

Vol.1.2, 1976, pp.75-90: 

Criteria for the Analysis of the Sutta Nipata 

The Sutta Nipata: Its Title and Form 

Vol.1.3, 1976, pp.136-63: 
The Vaggas of the Sutta Nipata 
Asoka's Bhabru Edict 
The Atthaka Vagga 

Other References in Buddhist Sanskrit Works 

The Chinese Arthapadam 

Astaka Varga or Arthaka Varga? 
Parayana Vagga 

Its Antiquity 

The Vatthu Gatha 
Uraga Vagga 
Culla and Maha Vaggas 

The Uraga Sutta 
The Khaggavisana Sutta 
The Muni Sutta 

Vol.2.2, 1977, pp.86-105: 

The Parabhava Sutta 

The Vasala Sutta 

The Mahamangala Sutta 

The Metta Sutta 

The Ratana Sutta 

Vol.2.3, 1977, pp.141-58: 
The Yakkha Ballads 

Hemavata Sutta 
Alavaka and Suciloma Suttas 
The Yakkha-legend (Alavaka) 
External Evidence 
The Pastoral Ballads 
Dhaniya Sutta 
Language and Syntax 

Doctinal Developments 
External Evidence 
Kasibharadvaja Sutta 

External Evidence 

The Narrative Ballads 
Pabbajja Sutta 
Padhana Sutta 
Nalaka Sutta 
External Evidence 
The Nalaka-dsicourse 

Vol.3.2, 1978, pp.45-64: 

Some Suttas from the Atthaga Vagga 

Kama Sutta 

The Atthakas 

Jara Sutta 

Magandhiya Sutta 
Pucchas from the Parayana Vagga 

Linguistic and other Internal Evidence 

The Other Puchas 

The Apadana and Sixteen Manavas 

Reference in other works 

Vol 3.3, 1978, pp.100-113: 

General Observations and Conclusions 

Vedic and Dialectic Variations 



are treated as independent articles. If there is no present form of a verb, 
the entry word is put in brackets. 

In the CPD proper names are also included, i. e. names of persons, 
localities, titles, and the like, each person, locality, title being treated as 
independent article. All indeclinable words, pronouns, numerals, original 
roots are treated independently. 

The names of the texts quoted are given in abbreviated forms, e. g. 
D (for Dighanikayd), M (for Majjhimanikaya), S. (for Samyuttanikaya), A 
(for Anguttaranikayd), Dp (for Dhammapadd), Mp (for Manor athapuran'i) 
Ps (for Papancasudani, Mhv (for Mahavamsa), Dhs (for Dhammasangant), 
Mhv-t (for Mahavamsa-tika), and the like. Different editions are 
indicated as E e (for Roman edition); C e (for Sinhalese edition); B e (for 
Burmese edition) S e (for Siamese edition). K e (for Cambodian edition). 
This apart, there are some typographical and other rules (regarding signs, 
symbols, technicalities, etc.) which we would not mention here to bore 
the readers of this article any more. 

Under the editorship of L. Alsdorf the CPD work has been running 
smoothly for the last few years. Professor Alsdorf is no doubt the most 
competent scholar for the work. But still there is no certainty when the 
CPD work will come to an end, and there is cent percent doubt if the 
work will at all come to an end in the present century. We should request 
the members of the Royal Danish Academy to find out means to expedite 
the progress of the CPD work so that it may come to an end within the 
span of the present century. We can suggest to find out scholars to work 
in their individual capacity, apart from the centre or centres, in different 
countries of the world and if possible, to set up more and more centres 
where the retired college and university teachers will work as whole- 
timers with the assistance of the whole-timer research assistants and 
collaborators. Of course this would be possible if suitable grants are 
available from UNESCO and other Foundations and Institutions. 



N. A. Jayawickrama 

(We are grateful to Prof. Jayawickrama for permission to serialise his 
doctoral thesis which was accepted by the University of London in 1947 
under its full title of "A critical analysis of the Pali Sutta Nipata illus- 
trating its gradual growth". It was subsequently serialised in the Ceylon 
University Review — 1943-67 — in its issues January 1948 to April 1951.) 


h—Anguttara Nikaya (PTS, 6 vols. 1885-1910, 1956-61) 

BSk— Buddhist Sanskrit 

DA— Digha Nikaya (PTS, 3 vols. 1889, 1903, 1910; 1967, 1967, 1960) 

H. O. S. — Harvard Oriental Series 

I.H.Q.— the Indian Historical Quarterly (Calcutta, 1925-63) 

M—Majjhima Nikaya (PTS, 4 vols. 1887-1925, 1960-74) 

Uiln—Milindapanha (PTS, 1880, 1963) 

Pj—Paramatthajolika II (^Suttanipata Commentary II, PTS, 1917, 1966) 

Pug— Puggalapanitatti (PTS, 1883, 1972) 

Pv— Petavatthu (PTS, 1889) 

S—Samyutta Nikaya (PTS, 6 vols. 1884-1904; 1960-73) 

S.B.E. — Sacred Books of the East series 

Sn— Suttanipata (PTS, 1913, 1965) 

SnA I— Suttanipata Commentary I (PTS, 1916, 1966) 

Th. \—Theragatha\ Th. 2—Therlgatha (PTS, 1883, 1966) 

The Criteria for the Analysis of the Sutta Nipata 


The Sutta Nipata contains older and younger material side by side. The 
Atthaka and the Parayana Vaggas preserve, on the whole, older composi- 
tions. Many suttas included in the other three vaggas too can be 
established, without doubt, to belong to an equally old, or perhaps older 

It is our present task to investigate whether the compilation of the 
Sutta Nipata (as a separate work) was done by gradual stages or was 
the work of a single editor. It is certain that at least its last two vaggas 
had a separate existence prior to their being incorporated in the Sutta 
Nipata, for there are numerous references to them in Pali, Buddhist 


Sanskrit and Chinese works, with no mention of the Sutta Nipdta at 
all. Parts of the rest of the vaggas too appear to have existed in separate 
groups, but the Sutta Nipdta, as it is preserved now, is a compilation of 
a comparatively later date. The lateness of the compilation has no bearing 
whatsoever on the date of its constituent suttas. Chalmers, in his transla- 
tion of the Sutta Nipata entitled, Buddha's Teaching in H.O.S. Vol., 37, 
p. xvi, remarks, "the ascertained stages of growth of a compiled 'book' 
by no means settle the relative date of composition of its contents, a 
question for solution of which internal evidence must be invoked, for 
what it is worth." The internal evidence which helps to establish the 
relative date of composition of the suttas is primarily linguistic, but this 
alone is not sufficient. A study of the contents of the Sutta Nipata along 
with its metre and style, doctrinal developments, and social conditions 
depicted in them will greatly supplement whatever information linguistic 
evidence yields. Whenever external evidence is available in support of 
internal evidence more definite results can be achieved. 

Linguistic evidence consists mainly of an analysis of words in their 
form and use, of tenses, of syntax and of vocabulary. As early as 1880 
Fausbbll (Translation to Sutta Nipata, S. B. E. vol. X, pp. xi. ff.) has 
pointed out, "We not only find here what we meet with in other Pali 
poetry, the fuller Vedic forms of nouns and verbs in the plural. . .the 
shorter Vedic plural and the instrumental singular of nouns. . .Vedic 
infinitives, . . .contracted (or sometimes old) forms, . . .by the side of 
protracted forms, but also some unusual (sometimes old) forms and 
words. . .We also find tmesis as in the Vedas. . .Sometimes we meet with 
difficult and irregular constructions, and very condensed expressions." 
He also notes that the parts of the Sutta Nipdta containing these 
"irregularities" are much older than the suttas in which the language 
is fluent and the verses are melodious. This practically covers the whole 
field of linguistic evidence that can be gleaned in the Sutta Nipdta. 

A comparison of the linguistic peculiarities of the various parts of the 
Sutta Nipdta with Vedic, the language of the Brdhmanas, Pali of the 
got/id-literature, Canonical prose, and Classical Sanskrit helps in some 
degree to fix the relative dates of the suttas. It has been already stressed 
that the importance of linguistic data should not be over-estimated, for, 
these alone without other corroborative evidence are not of very great 
value. More definite conclusions can be drawn when they are supported 
by other internal and external evidence. 



Other internal evidence consists of metre and style, doctrinal develop- 
ments and ideology, and social conditions. As a rule, metre is no proper 
criterion of judgment in assigning relative dates to Pali poetry. The 
majority of the metres employed in Pali is to be met with in earlier 
literature both Vedic and early Sanskritic. The developments and modify 
cations that earlier existing metres have undergone in Pali may lead to 
some valuable information; but such changes invariably have their 
parallels in earlier Sanskritic literature. This minimises the importance 
of any evidence from this source. The changes in metre from which 
somewhat definite inferences could be drawn are to be met with only 
in very late Pali poetry; e.g., the Ceylon Chronicles. 

The most popular metre in the Sutta Nipdta is Anustubh Sloka. There 
are no less than 562 stanzas in anustubh metre, in addition to 54 modified 
anustubh slokas in the Vatthu-gdthd of the Pdrdyana, making a total of 
616 stanzas. {Vide Helmer Smith: "Metres of Sutta Nipata". Pj. II. 3, 
pp. 637-644). Next comes tristubh metre, which is employed in 374, 
stanzas. There are also 29 stanzas in drya metre, and 117 in vaitdliya 
and its allied metres, aupacchandasika and vegavati. Of these 117 stanzas 
only 15 are in pure vaitdliya, 41 are in aupacchandasika, 16 in vegavatl 
and the other 45 in mixed vaitdliya. 

Chalmers, (ibid, p. xvii) maintains that anustubh is later than tristubh 
and quotes the example of the four Atthakas in tristubh metre preceded 
by the Kama Sutta in anustubh sloka, stating that it ' 'manifestly forms 
a late preface to the Atthaka Vagga as a re-edited whole." He notes 
the change of metre in Sariputta Sutta and remarks that "the equally 
edifying slokas Nos. 955-62 suggest an editorial preamble to the vigorous 
tristubhs with which the Atthaka Vagga ends." He refers to the only 
Tristubh verse in Dvayatdnupassand Sutta (Sn. 728), and the tristubhs 
that are freely distributed in the Pdrdyana as being much older than 
the rest of the stanzas in those sections which he calls ' 'scholastic accre- 
tions." He advances another hypothesis that "the longer the metrical 
line the later is the composition likely to have been." (ibid). 

Keith {A History of Sanskrit Literature, p. 417), too, believes that 
the longer metrical line is a later development in Classical Sanskrit. 
Thus, it may be possible, purely on theoretical grounds, that those verses 


i I 

of the Sutta NipSta in arya, vaitaliya, aupacchandasika, vegavati and 
mixed vaitallva metres belong to a later stage of composition. Yet, there 
is no reason to assume that all the stanzas in the historically earlier 
tnstubh and anustubh metres are anterior to those written in later metres. 
Unlike other metres ana and vaitaliya are measured by the number 
of morae. {Vide Macdoncil, A Vedic Grammar for Students, p. 436 n. ->■ 
These metres in which the sum-total of morae was absolutely fixed 
probably developed from popular poetry according to Keith (op. at. 
p 418) 'and belons to the Classical epoch. Thus, Chalmers 1 hypothesis 
is generally applicable to the verses of the Sutra NipSta though he is 
not quite correct in the application of his hypothesis to tnstubh and 
anustubh metres. 

Both tnstubh and anustubh metres can be traced back to Vedic. About 
40 per cent of the stanzas in the Rgveda are written in trisfubh , whereas 
anustubh constitutes only about 8 to 9 per cent. (Vide Macdonel o, aL 
PP 438 ff.). Though the former is very popular in the Rgveda - the latter 
is the most predominant metre in the post-Vedic period (Macdonell 
ibid) Thus, generally anustubh slokas in the Sutta Nipata may be expected 
to be later than the tnstubh verses. Here, the hypothesis regarding tne 
length of the metrical line is inapplicable, as tristubh which is considered 
older has a longer metrical line (4X11) than anustubh (4X8). Moreover, 
over 86 per cent, of the stanzas in the Sutta Nipata are written m these 
two metres, and the number of stanzas written in other metres is just 
under 14 per cent. This being the case, Chalmers' suggestion, however 
true it may be, is of no great practical importance. 

The historical order of these metres occurring in the Sutta Nipata 
would be 1. tnstubh and anustubh, 2. and, vaitaliya, aupacchandasika, 
vegavati and mixed vaitaliya. There is no guarantee whatsoever that 
stanzas written in older metres are necessarily earlier than those m later 
metres Therefore, metre by itself is no sound criterion ior fixing relative 
dates and it only forms a very useful source of confirmatory evidence. 

Style like metre, is closely allied to linguistic evidence As the Sutta 
NipL is not a homogeneous work, its style vanes in its different sections 
Its poems range from simple popular ballads like he Dhamya and 
Kaslhdradvaja Suttas to scholastic compositions hke the D„upas- 
unvl Sutta It also contains simple narratives like the Pabbajja and 
Pa dh!na Suttas or the Vatthu^tha of the Naiaka Sutta and Parayana 
Vag,a as well as drogue-ballads of various types besides didactic 
poems like the Kimslla or Dhanunaeariya Suttas in which the editorial 


hand is keenly felt. 1 A simple and easy style unhampered by poetic 
embellishment's, excessive ihythm and metrical perfection suggests an 
early composition rather than a later one. The use of excessive all iter a-^ 
lion, assonance, and slesa (word-play) and ail such accompaniments of 
a 'heavy style" is generally a sign of lateness. The use of such poetical 
devices "is greatly limited in the sections of the Sutta Nipata, which from 
other evidence can be classed as very early. 

The oft recurring refrain belongs to the field of popular poetry c't all 
periods. It is also probable that the ballads in which the dialogue element, 
predominates (e.g., suttas like the Dhaniya and Hemavata: and not the 
quasi-dialogue ballads in which an interlocutor asks a question and the 
Buddha is seen replying with a long uninterrupted discourse), were 
dramatised and became widely popular. These two facts do not lead to 
any clue regarding the relative dates of poems, but it could be noticed 
that style goes hand in hand with metre to support linguistic data, and 
that it is very useful as a criterion for fixing relative dates for these 
ballads . 

The form in which these suttas are found (viz., entirely in verse, or 
mixed verse and prose, etc.) is sometimes helpful as a criterion. 

Doctrinal developments, generally, are a good index to the time of 
composition of individual sections, rather than of a work as a whole. 
This is true of the majority of the works of the Pali Canon, as they con- 
tain material drawn from more than one stratum. No well-defined 
developments as such are to be noticed in the older ballads of the Sutta 
Nipata, but a gradual change can be marked in the later ones. Some 
fundamental concepts already found in the earlier ballads and other 
early literature are seen undergoing a gradual crystallisation in the 
later ballads. New ideas are also seen finding their way. One such 
instance is the concept of vasana (which will be discussed later on). 

Closely connected with doctrinal developments is the growth of ideas. 
In as much as the doctrinal emphasis lay on the earliest t enets of 

_ 1 There areat least 6 sulfas (viz. lheHiri-\ Dhammacariya-, KimsVa-, Utthana,- Subha 
situ- and Attadanda) which derive their names from their opening words. The suttas 
that are named after' a word or simile in the body of the text are more numerous There 
are 11 suttas (vi* dte Alavaka,- MimL- Hiri-, Kimslla,- Rahula, — theVathu-gatha are 
in irregular amMiibh-SiMdsita-, Kokaliya,- Vathu-gatha of Naiaka-, Dhotaka pucchS 
Todeyya pucchti, and Jatukanni puccha) in which the opening lines are written m a 
different metre from that (or those) of the rest of the poem. In five of these the opening 
stanza (or stanzas) is in anustubh. Less numerous are the poems in which the concluding 
stanzas are written in a metre different from that oj the rest of the poem: e. g. the 
Dhaniya-, Sahhiya-, Vahgisa-, Sundarikabharadvaja and Pasura Suttas. 


v . 1 HIM I k . JH WTPWWBBI WW. I M^^ «ui*hiwim w i mhhi i wm i iii »■ w ma ■ , j f . '■ ■■ ■» r 

- M^qfr-.-^v-TC . T A\«IK)«ai 



Buddhism — which Mrs. Rhys Davids prefers to call "Sakya" — in the 
majority of these ballads, so also could be noticed the gradual formation 
of definite ideas and concepts which in course of time came to play an 
important role in later Buddhism. Along with this appear standard 
technical expressions which too in course of time became fixed. Some 
terms are seen in the transitional stage of being crystallised in these 
ballads. The later ballads mark the gradual drift from primitive "Sakya" 
to monastic Buddhism which replaced it. The trend of development, 
if successfully traced, will enable one to place these suttas in some sort 
of chronological arrangement. 


Social conditions depicted in the Sutta Nipata reflect an age when 
Brahmanism held sway and caste exerted great influence. The ballads 
show that in spite of the effort of the Buddha to break down these barriers 
he was obliged to give new values to what was best in Brahmanism; 
(e.g., the Buddhist connotation of brahmana, etc.), in order to make his 
message universally acceptable. Society was mainly agricultural and 
there lived rich herdsmen like Dhaniya (a Vaisyal) and brahmin farmers 
like Bharadvaja. The samanas and paribbajakas are accepted institutions 
and many paribbajakas are represented as getting their individual 
problems solved by the Buddha. 

The older ballads reflect a time when Buddhism had not developed 
into a full-fledged monastic (coenobitic) system. It is the muni, the 
bhikkhu, or the samana, that these ballads are concerned with. There 
are only two references to thera in the whole of the Sutta Nipata, both 
occurring in introductory prose at pp. 59 and 92 respectively. The latter 
reference is not to Buddhist theras, but to those who are "firmly estab- 
lished in their own religious beliefs. ' ' The conditions among the bhikkhus 
were most probably far different from those prevalent during the time 
of the composition of the Thera-aad Theri-gdthas. There appears no 
organised monastic body; but on the contrary there were the munayo 
(ascetics in general) or the bhikkhus who were expected to lead the life 
of a muni. 

The social conditions reflected in the Sutta Nipata regarding peoples 
and castes, countries and towns, brahmins and sacrifice are no different 
from those reflected in the prose Nikayas. It is probable that the majority 
of the Pali works generally depict conditions prevalent at the time of 
their composition, but the difference of a century or two hardly makes 
any fundamental difference in the structure of society and mode of life 
in those far-off days. 


Incidental references to contemporary history would enable one to 
draw some conclusions regarding the time of composition. Often such 
references are not made directly. They occur as anachronisms. One 
such instance is to be noticed in mandira—a. political division; which 
probably came into being after the formation of a large empire. Thus, 
any evidence gathered from this source too will be seen to supplement 
what has already come to light from other sources. 


External evidence is of utmost importance. Several Canonical works 
make reference to, and quote from certain suttas and sections in the 
Sutta Nipata. This necessarily proves that the sections of these works 
which refer to and quote from the Sutta Nipata are decidedly later than 
those respective suttas of the Sutta Nipata. The references made to the 
Atthaka and the Parayana Vaggas will be discussed later. Equally 
numerous are the references made to these sections in the later BSk. 
and Chinese Buddhist literature. The Atthaka Vagga occurs in full in 
Chinese (i.e., No. 198 Thai Shu Tripitakd). Besides these references in 
literature there is important inscriptional evidence in Asoka's Bhabra 
edict. All these external data are connected with individual suttas and 
there is no specific mention of the Sutta Nipata in any early work. It 
is mentioned for the first time in the Milinda Panha. 


Another criterion is the indirect evidence from the position of the 
suttas as they occur in the vaggas. Some suttas are placed at the head 
of the vagga for their outstanding merit (e.g., the opening suttas of the 
Uraga Vagga) while other opening suttas bear definite signs of lateness 
(e.g., Ratana). Of equal importance are the suttas occurring at the end 
of the vagga. The Muni Sutta, in spite of its being an old piece is placed 
at the end of the Uraga Vagga after a relatively younger piece Vijaya 
Sutta. On the other hand, the late Dvayatanupassana Sutta concludes 
the Maha Vagga. 

In the light of all these conflicting data it is not possible to formulate 
a working principle to be guided by. However, it will be seen that some 
of these suttas are younger in time and in general tone. When older 
suttas in similar positions are also taken into account these younger 
pieces point to a redaction of the suttas subsequent to an earlier colla- 
tion rather than to their being interpolations. 

A striking similarity is to be seen in the Vinaya. The popular tradition 
has been embodied in the opening chapters of the Maha Vagga, while 


nikaya it is comparatively late, and is much later than the other four 
Nikayas (vide Winternitz, op. cit. Vol. II, pp. 77 fT). The Sarvastivadins 
speak of only the four earlier Nikayas (which they call Agamas), and 
the Theravadins themselves had reached no general agreement regarding 
the number of works in the Khuddaka Nikaya. Even at the present day 
in countries where Pali Buddhism prevails the number of books which 
should constitute this nikaya is not agreed upon. 


The Jatakas which form the tenth book of the Klmddaka Nikaya 
in the Theravada Canon are also considered as a separate anga (lit, 
limb, i.e. division) in the descriptive classification known as the navanga- 
satthusasana (the Ninefold Dispensation of the Teacher) which occurs 
in many places in the Canon itself e.g. M. I, 130; A. II, 103, 178, III, 
86 &.; Ill ff., Pug. 43, Miln. 344 etc. Although ihis classification is 
necessarily old (vide E. J. Thomas: Life of Buddha p. 167, where he 
considers the division into angas as earlier than that into nikayas) it 
does not speak of any definite works, for, a jataka may be included in 
a sutta, an udana in a veyyakarana etc. Like the Udana and the Itivuttaka, 
the name Jataka coincides with that of an actual work in existence. 
But there is nothing to say that by this angaw&s meant the present Jataka. 
As Dr. E. J. Thomas (History of Buddhist Thought, p, 227) says "the 
probability is that the terms were used. , .to describe the character of 
the composition" rather than signify actual works. He points out that 
there are numerous instances of udanas and jatakas in various parts of 
the Canon which are not included in the works known by these respec- 
tive names; (ibid) c\g. ? the Mahasudassana Sutta in the Digha Nikaya, 
jatakas m Cariya Pitaka, Sivi Jataka called Sivi Sutta in Miln^ etc. 
So is also the case with the hivuttaka, 

Taking up the division of the Pali Navanga, its first ahga, Sutta, is 
said to include the Yinaya, certain suit as in gatha, and other sayings of 
the Buddha classed as sutta. It is noteworthy that Commentaries con- 
sidered these sections of the Sutta Nipata which did not fall into the 
category of sutta, as gatha, the fourth, anga {vide DA. I, 23); but gatha 
primarily consisted of verses in Dh. Th. 1 and Th. 2 (vide Thomas, op. 
cit.). The Commentary says that the Sutta Nipata consists of gatha. (verse), 
geyya (mixed prose and verse) and veyyakarana (expositions) which, 
on account of their informative, instructive and expository nature are 
called suttas, and that the work is called the Sutta Nipata because it 
contains such suttas grouped together (SnA. \ — Pj. II introduction). 
From these statements it is clear that at least some suttas, if not the 


majority of them in the Sutta Nipata. can be said to belong to the Sutta 
Anga (vide Thomas, op. dr.). 

Of the known instances of nipdtas in the Pali Canon, the Ahguttara 
as a work has nothing corresponding to it in the Navahga division, the 
Jataka may have been considered to correspond to the seventh anga, 
Jataka, and it is probable that the Sutta Nipata was only a nipata of a 
similar anga. This only implies that the Sutta Nipata consists of some 
suttas representative of the type Sutta and therefore is a nipata of suttas. 


This collection should consist entirely of pieces which could be desig- 
nated as Sutta if the title Sutta Nipata were to be justified. The Com- 
mentary (SnA.) states that the three types gatha, geyya and veyyakarana 
can be again called sutta and therefore the gatha in the Sutta Nipata 
are suttas as well. It is not possible to draw a line of demarcation between 
gatha and sutta. Of the 72 pieces found in the Sutta Nipata as many as 
54 i.e. those forming vaggas I-IV, are called sutta by name, irrespective 
of whether they would strictly be categorised as sutta or gatha, if such 
a division were possible. (The other 18 pieces form the Pardyana con- 
sisting of the prologue— vatthu- gatha, the 16 pucchds and the epilogue 
respectively). This fact probably furnishes a clue to this problem. 
During the time of the arrangement of this collection the distinction 
between gatha and sutta may not have been strictly observed, and things 
may have existed in a rather fluid state. 

Id the first 54 pieces a growing tendency towards standardisation can 
be seen. Every piece, whether ballad or discourse, is termed a sutta. 
The stanzas of the so-called suttas are often referred to as gatha: e.g. 

Sn. 429 ed, 

Ima gatha bhanam maro attha Buddhassa santike 
(Uttering these stanzas Mara stood near the Buddha); Sn. 251c, 

citrahi gdthdhi muni-ppakasayi 
(the sage declared in colourful verse) in the narrative section of the 
Amagandha Sutta; Sn. pp. 13, 32, 46 and 48 in the narrative prose of 
the Kasibhdradvaja, Alavaka, Mahamahgala and Suciloma Suttas 


Atha kho. . . .Bhagavantam gdthdya ajjhabhasi. (Then indeed, N. N. 
addressed the Bhagava in a stanza); Sn. p, 78, 

Saruppahi gdthdhi abhitthavi (extolled him with appropriate stanzas) 
in the prose of the Subbhasita Sutta. It also occurs at Sn. 81=480 in the 
phrase, gdthdbhigiiam (what is obtained by reciting stanzas) and Sn. 
1131 a, parayanam amtgayissam (I shall sing the Parayana). 


nikaya it is comparatively late, and is much later than the other four 
Nikayas (vide Winternitz, op. cit. Vol. II, pp. 77 ff.). The Sarvastivadins 
speak of only the four earlier Nikayas (which they call Agamas), and 
the Theravadins themselves had reached no general agreement regarding 
the number of works in the Khuddaka Nikaya. Even at the present day 
in countries where Pali Buddhism prevails the number of books which 
should constitute this nikaya is not agreed upon. 


The Jatakas which form the tenth book of the Khuddaka Nikaya 
in the Theravada Canon are also considered as a separate anga (lit. 
limb, i.e. division) in the descriptive classification known as the navanga- 
satthusdsana (the Ninefold Dispensation of the Teacher) which occurs 
in many places in the Canon itself e.g. M. I, 130; A. II, 103, 178, III, 
$6 ff.; 177 ff., Pug. 43, Miln. 344 etc. Although this classification is 
necessarily old (vide E. J. Thomas: Life of Buddha p. 167, where he 
considers the division into angas as earlier than that into nikayas) it 
does not speak of any definite works, for, a jataka may be included in 
a sutta, an uddna in a veyyakarana etc. Like the TJdana and the Itivuttaka, 
the name Jataka coincides with that of an actual work in existence. 
But there is nothing to say that by this anga was meant the present Jataka. 
As Dr. E. J. Thomas (History of Buddhist Thought, p. 227) says "the 
probability is that the terms were used. . .to describe the character of 
the composition" rather than signify actual works. He points out that 
there are numerous instances of uddnas and jatakas in various parts of 
the Canon which are not included in the works known by these respec- 
tive names; (ibid) e.g., the Mahasudassana Sutta in the Digha Nikaya, 
jatakas in Cariyd Pitaka, Sivi Jataka called Sivi Sutta in Miln., etc. 
So is also the case with the Itivuttaka. 

Taking up the division of the Pali Navanga, its first anga, Sutta, is 
said to include the Vinaya, certain suttas in gatha, and other sayings of 
the Buddha classed as sutta. It is noteworthy that Commentaries con- 
sidered these sections of the Sutta Nipdta which did not fall into the 
category of sutta, as gatha, the fourth anga (vide DA. I, 23); but gatha 
primarily consisted of verses in Dh. Th. 1 and Th. 2 (vide Thomas, op. 
cit.). The Commentary says that the Sutta Nipdta consists of gatha (verse), 
geyya (mixed prose and verse) and veyyakarana (expositions) which, 
on account of their informative, instructive and expository nature are 
called suttas, and that the work is called the Sutta Nipdta because it 
contains such suttas grouped together (SnA. 1 — Pj. II introduction). 
iFrom these statements it is clear that at least some suttas, if not the 


majority of them in the Sutta Nipdta, can be said to belong to the Sutta 
Anga (vide Thomas, op. cit.). 

Of the known instances of nipatas in the Pali Canon, the Ahguttara 
as a work has nothing corresponding to it in the Navanga division, the 
Jataka may have been considered to correspond to the seventh anga, 
Jataka, and it is probable that the Sutta Nipdta was only a nipdta of a 
similar anga. This only implies that the Sutta Nipdta consists of some 
suttas representative of the type Sutta and therefore is a nipata of suttas. 


This collection should consist entirely of pieces which could be desig- 
nated as Sutta if the title Sutta Nipata were to be justified. The Com- 
mentary (SnA.) states that the three types gatha, geyya and veyyakarana 
can be again called sutta and therefore the gatha in the Sutta Nipdta 
are suttas as well. It is not possible to draw a line of demarcation between 
gatha and sutta. Of the 72 pieces found in the Sutta Nipdta as many as 
54 i.e. those forming vaggas I-IV, are called sutta by name, irrespective 
of whether they would strictly be categorised as sutta or gatha, if such 
a division were possible. (The other 18 pieces form the Pardyana con- 
sisting of the prologue — vatthu-gatha, the 16 pucchas and the epilogue 
respectively). This fact probably furnishes a clue to this problem. 
During the time of the arrangement of this collection the distinction 
between gatha and sutta may not have been strictly observed, and things 
may have existed in a rather fluid state. 

In the first 54 pieces a growing tendency towards standardisation can 
be seen. Every piece, whether ballad or discourse, is termed a sutta. 
The stanzas of the so-called suttas are often referred to as gatha; e.g. 
Sn. 429 ed, 

Ima gatha bhanam mdro attha Buddhassa santike 
(Uttering these stanzas Mara stood near the Buddha); Sn. 251c, 

citrahi gdthdhi muni-ppakdsayi 
(the sage declared in colourful verse) in the narrative section of the 
Amagandha Sutta; Sn. pp. 13, 32, 46 and 48 in the narrative prose of 
the Kasibharadvdja, Alavaka, Mahamangala and Suciloma Suttas 

Atha kho. . . .Bhagavantam gathaya ajjhabhdsi. (Then indeed, N. N. 
addressed the Bhagava in a stanza); Sn. p, 78, 

Saruppahi gdthdhi abhitthavi (extolled him with appropriate stanzas) 
in the prose of the Subbhdsita Sutta. It also occurs at Sn. 81=480 in the 
phrase, gdthdbhigllam (what is obtained by reciting stanzas) and Sn. 
1131 a, parayanam anugdyissam (I shall sing the Pardyana). 



Besides these there are three instances of introductory verses called 
Vatthugatha viz. 1. A short introduction to the Rahula Sutta (Sn. 335-336), 
2. the introduction to the Nalaka Sutta (Sn. 679-698) and 3. the prologue 
to the Parayana (Sn. 976-1031). 

Again in the Bhabru (or Bairat) Minor Rock Edict of Asoka (vide 
Hultzsch: (Corpus Inscriptionum Indicarum, Vol. 1, p. 172) the fourth 
dhammapaliydya (section of the Scriptures) which is identified with the 
Muni Sutta of the Sutta Nipata (vide Dharmananda Kosambi: Indian 
Antiquary 1912 pp. 37 ff.) is called Muni-gatha (Stanzas on a Sage). 
The sixth which is identified with the Sariputta Sutta of the Sutta Nipata 
(ibid.) is called Upatisapasine (the question of Upatissa). Here too is 
noticeable the tendency towards standardisation, for, what were known 
to Asoka as gatha and pasine are called suttas in the Sutta Nipata. Thus, 
it can be seen that an attempt has been made to designate as suttas, 
as many pieces of the Sutta Nipata as possible. Hence the compiler 
has found no difficulty in classifying this work as a nipata in a larger 
group of suttas. It is very unlikely that he had the Sutta Pitaka in mind, 
and it is quite probable that the Sutta Nipata was meant to be a nipata 
among suttas in general, if not in the ahga of Sutta, although finally it 
came to be considered as a nipata of the Khuddaka Nikaya. 

The Form of the Sutta NipSta 

The suttas of the Atthaka Vagga and the Pucchas of the Parayana 
Vagga are entirely in verse, whereas suttas of the other three vaggas 
are of two different types, one in pure verse, the other partly in verse 
and partly in prose. The 20 suttas in vaggas 1-111 which are entirely 
in gatha form and called "Verse Ballads" (vide, S. M. Katre: Early 
Buddhist Ballads and their Relation to Older Upanishadic Literature) 
are distributed in the following manner: 7 in Uraga Vagga, 9 in Culla 
Vagga and 4 in Maha Vagga. The "Mixed Ballads" (in prose and verse) 
occur as 5, 5 and 8 in the three respective vaggas. The absence of Mixed 
Ballads in the Atthaka and Parayana Vaggas and the ascending order 
in which they occur in the other three vaggas may furnish valuable data 
in discussing the relative chronology of these suttas. 

The prose in these suttas is not an essential factor in the dialogue or 
discourse as the case may be. It is employed as an aid to the narrative 
or to describe the nidana (the context). The only prose in seven 1 of these 
eighteen Mixed Ballads is the passage describing the circumstances 
Alavaka, Mahdmangala, Suciloma, Vaiiglsa, Dhammika and 


eading up to the dialogue or discourse in verse. Six suttas' 1 contain an 
additional prose passage, following the verse, which is very similar to 
one another in five instances describing the confession of faith by the 
Buddha's interlocutors. The other five suttas 3 contain three or more 
prose passages many of them interspersed with the verses. 


A close examination of the prose passages shows that they did not, 
as a rule, form an essential part of the ballad to begin with. In four of 
the last category of suttas i.e. except the Dvayatanupassana, it serves 
merely as a connecting thread running through the whole ballad linking 
up the various parts. Generally, when what is stated in the stanzas is 
not sufficient for the listener to grasp what has transpired between the 
end of one part and the beginning of the next part of the ballad, prose 
is introduced giving the necessary details; e.g. Sn. p. 14, 

Atha Kho Kasibharadvajo brdhmano mahatiya kamsapatiya pdydsam 
vaddhetva Bhagavato upanamesi (Then the brahmin K. served out milk- 
rice in a large bronze bowl and offered it to the Bhagava). Also see Sn. 
p. 110. Sometimes with the change of speaker prose is introduced; e.g. 
Sn. p. 79, in the Subhasita Sutta; and often for both the above reasons; 
e.g. Sn. p. Ill Sela and pp. 94-100 Sabhiya Suttas. 

The language of the prose is quite similar to that of the prose Nikayas 
in idiom, syntax and style. The stereotyped expression in the prose of 
the Sutta Nipata does not permit one to infer that it preserves the exact 
words of the narrators or reciters of these ballads. Generally, ballad- 
reciters state in their own words, such facts as are necessary for the 
listeners to follow the narrative in the ballads. Here the prose states 
the same facts though clothed in the standard Canonical garb; and 
probably this standardisation has taken place long after the composition 
of the ballads themselves. 

Narrative prose should be compared with verse employed for narra- 
tion, found in abundance in the Sutta Nipata. The Commentator himself 
attributes several stanzas to the sangltikara (reciters at a Samiglti or 
"Council", i.e., compilers); e.g. Sn. 30, 251-252, 355d, 401d, 429cd, 
449 and the epilogue of the Parayana at (SnA. 42, 292, 351, 377, 387, 
394 and 603 respectively). In addition to these he attributes to the 
satigitikara, such repertory phrases as, iti Bhagava (Sn. 355 etc.), iti 
brdhmano (Sn. 459 etc.) and prose elements in the Sabhiya, Sundarika- 
bharadvaja, Magha, Sela and Dvayatanupassana Suttas (at SnA. 351, 


2. Vasala, Brahmanadhammika, Sundarikabharadvaja, Magha and Vaseffha. 

3. Kasibharadvajo, Subhasita, Sabhiya, Sela and Dvayatanupassana. 


405, 394, and 398, 400, 414, 456 and 504 respectively). The vatthu-gatha 
of the Nalaka Sutta and Parayana Vagga also belonged to the sangltikara 
according to the Commentary (SnA. 483 and 580 respectively). 

On a broad basis, the language, metre and style of the passages which 
are attributed to the sangltikara are no different from those of the other 
parts of the ballads to which they belong, for, their language, like that 
of the rest of the gathas in the Sutta Nipata preserves an earlier phase 
of Pali than the standard Canonical expression of the prose of the Sutta 
Nipata. (Also vide Geiger, Pali Literatur und Sprache, p. 1.). It is quite 
probable that in most cases this "narrative element" in verse goes 
back to the time of the composition of the ballads themselves. 

On the other hand, the narrative prose in its present form cannot, in 
any way, date back earlier than the period when the Canonical prose 
idiom was gradually being fixed and acquired an accepted standard 
form. It is not improbable that this prose dates back only to the time of 
the arrangement of the Sutta Nipata as a separate work. Prior to that 
time no fixed prose narrative may have been attached to these ballads, 
and the reciters used their own words when necessary. Thus, the prose 
in the Sutta Nipata can be considered as being much younger than the 


The poetical pieces in the Sutta Nipata are of three main types: — J.. 
Simple didactic verse, 2. Dialogue in verse and 3. (Didactic) discourse 
or dialogue following a prose introduction. Type 2 can be further sub- 
divided into (a) pure dialogue in verse, (b) dialogue consisting of a 
discourse in answer to a question. 

There are 21 suttas belonging to type I, viz. I, 1, 3, 8, 11, 12; II. 1, 3, 
6, 8, 10; III. 8; IV. 1-6, 8, 12, 13 and 15. Some of these suttas like I. 1 
(Uraga) and I. 3 (Khaggavisana) etc. are simple ballads with a regular 
refrain running through them. Others like I. 8 (Metta), II. 1 (Ratana) 
and II. 3 (Hiri) etc. dilate on certain topics of religious or doctrinal 
importance; still others such as some of the suttas from the Atthaka 
Vagga (included in the above list) show the attitude of a true follower 
of the Buddha to certain then-current issues. The last two sub-types 
are more in the nature of discourses rather than simple ballads. 

There are 30 pieces belonging to type 2; 20 of which viz. I. 2, 5, 9; 
IV. 9 and V. 2-17 (the sixteen pucchas) can be said to belong to type 2(a) 
i.e. dialogues in verse. Class 2 (b) consists of the 10 suttas, II. 2, 9, 11, 
13; III. 11; IV. 7, 10, 11, 14 and 16 in which a discourse in verse is given 
in reply to a question asked by an interlocutor. 

Type 3 consisting of the so-called "Mixed Ballads" includes 16 suttas; 
viz. I. 6, 7, 10; II. 4, 5, 7, 12, 14; III. 3-7, 9, 10 and 12. Some of the suttas 
like I. 7 (Vasala), II. 4 (Mahamangala), III. 3 (Sundarikabharadvaja), 
III. 5 (Magha) etc. are discourses in the form of ballads; while others 
like I. 6 (Parabhava) I. 10 (Alavaka) and II. 12 (Dhammika) are dialogues 
on matters of doctrinal importance. 

The other 5 pieces which are not included in the above classification 
are I. 4 (Kasibharadvaja S.), a prose and verse mixed narrative with 
dialogue, V. 18 (the epilogue to the Parayana) a prose and verse mixed 
narrative, and III. 1 (Pabbajja S.) III. 2 (Padhana S.) and V. 1 (the Vatthu- 
gatha of the Parayana, simple narratives in verse.) 


The ballads of the Sutta Nipata are popular in character, though they 
describe incidents connected with the Buddha and his teaching. There 
is a great deal of popular lore incorporated in the gathas e.g. Sn. 137-141, 
667-678, etc. There are also many popular teachings in some of the suttas, 
e.g., 1, 6, 7, 8 and the late Ratana Sutta (II. 1); but all of them are at 
the same time characteristically Buddhist. The ballads also contain 
many passages and ideas common to the earlier Upanisads and the 
Epics; {vide Katre, op. cit.). 

Besides these similarities that the Sutta Nipata bears to the earlier 
Upanisads and epic literature, it has much in common with the earlier 
Sanskritic literature even in form. The narrative-ballads, viz. Pabbajja, 
Padhana and Nalaka (vatthu-gatha only) Suttas have their counterpart 
in the akhyana (ballad) literature in Sanskrit. Their common characteristic 
is the alternation of dialogue stanzas with narrative stanzas. Discussing 
these suttas Winternitz (op. cit., Vol. II, p. 96) remarks that they are 
"precious remnants of that ancient sacred ballad-poetry from which 
the later epic version of the life of Buddha grew, in the same way as 
the heroic epic grew out of the secular ballads or akhyanas. 

In the same way, the riddle poetry found in the Sutta Nipata, such 
as the Alavaka and Hemavata Suttas in which a yakkha asks a question 
has parallels in the Mahabharata (vide Winternitz, ibid. Vo. I, p. 352 
and P.V. Bapat, The nagari edition of the Sutta Nipata, p. XVII). The 
poetical riddles or brahmodya of the Rgveda, e.g. I. 164, VIII. 29 are 
not very different from the riddle-poetry of the Sutta Nipata (cp. Kasi- 
bharadvaja S.). The mixed prose and verse narrative dialogues of the 
Brahmanas are an exact parallel to the "Mixed-Ballads" of the Sutta 
Nipata. From these it is evident that the early Buddhists not only used 
the same traditional floating literary material, but also made use of 
the same literary modes common to the earlier Sanskritic literature. 


The dramatic element which is not rare in the Sutta Nipdta has its 
parallels in the earlier literature. It is clearly noticeable at I. 2. (Dhaniya 
S.), I. 9 (Hemavata S.), III. 2 (Padhdna S.) and IV. 9 (Magandiya S.). 
It may be said that the majority of the dialogue ballads can be dramatised; 
but in the absence of any positive evidence it cannot be established with 
certainty that any of them were dramatised in early times. There is only 
a certain degree of probability. 4 

There is no doubt that these ballads were sung. The internal evidence 
of the Sutta Nipdta itself testifies to it; e.g. Sn. 81=480 gathdbhigitam 
(what is obtained by singing stanzas), Sn. 682a, Selentigdvanti ca vddiyanti 
(they cry exultantly, sing and play instrumental music). As suggested 
by Katre (op. cit.) it is probable that these stanzas were sung to the 
accompaniment of music (cp. Sn. 682a); but the only evidence he puts 
forward is the occurrence of the word rind (lute) at Sn. 449b, rina kaccha 
abhassatha (the lute dropped from under his arm lit. arm-pit). Judging 
from the fact that a vlna (which is usually associated with his three 
daughters) was incongruous with the early Maia-legend and that the 
parallel line in Mahdvastu reads, vindsam gacchi ucchriti (His pride was 
all shattered — Mavastu. II. 240) much significance cannot be attached 
to this line. However, the very form of the gdthas suggests that they 
were sung, and it is probable that the regular dialogue ballads were 
sung on suitable occasions (samajjas?) by two or more reciters, each 
singing the respective words spoken by the characters in the ballad. 

In the Dhaniya Sutta for instance, two reciters would sing the alternate 
stanzas representing the dialogue between the herdsman Dhaniya and 
the Buddha, a third would introduce Sn. 30, the words of the narrator, 
while Mara appears singing Sn. 33. Here is a regular dramatic piece. 
Such ballads can be compared with the akhydna-hymns, of the Rgveda 
(e.g. the hymn about Sarama and the Panls, Rv. X. 108, or the dialogue 
between Yama and YamI, Rv. X. 10), which are regarded by some as 
the earliest forms of dramatic literature in India and by others as ballads 
(vide Ghate's Lectures on Rgveda, p. 121 n. 1). In fact, the dkhydna- 
hymns of the Rgveda, on account of their dual characteristic of being 
ballads and dramatic pieces at the same time, can be said to bear a very 
close resemblance to the dialogue ballads of the Sutta Nipdta. 


There are numerous references in the Nikayas to dramatic performances, e.g., 
na(a, nacca, visukadassana, pekkha, samajja and sobhanika : vide O. II. de A 
Wijesekera, "Buddhist Evidence for the Early Existence of Drama.''' I.H.Q. 
XVII, where he has analysed the data giving many references. It is probable that 
Buddhist ballads were dramatically recited at Samajjas and similar occasions. 






Bhikkhu Bodhi 


The Buddha's Teaching is concerned with a single problem, the problem 
of dukkha or suffering, and the task it imposes is likewise of a single 
nature— the task, namely, of bringing dukkha to an end. 

In the standard formulation of the Four Noble Truths, the Buddha 
defines the truth of dukkha, the first Noble Truth, thus: 

"What, monks, is the Noble Truth of Dukkha? Birth is dukkha, 
decay is dukkha, death is dukkha; sorrow, lamentation, pain, dis- 
pleasure and despair are dukkha; union with the unpleasant is dukkha, 
separation from the pleasant is dukkha, not to get what one wants is 
dukkha; in brief, the five aggregates of clinging are dukkha. This, 
monks, is the Noble Truth of Dukkha." 1 

The five aggregates of clinging (pancupdddnakkhandhd) present a 
complete epitome of dukkha, both extensively by way of range and 
intensively by way of essence. Since this is so, we sometimes find that 
the formula for the first truth deletes the specific instances of dukkha 
and defines its subject matter directly as the aggregates: 

"What, monks, is the Noble Truth of Dukkha? The answer is: the 
five aggregates of clinging; that is, the clinging aggregate of material 
form, the clinging aggregate of feeling, the clinging aggregate of 
perception, the clinging aggregate of volitional determinations, and 
the clinging aggregate of consciousness. This, monks, is the Noble 
Truth of Dukkha:' 2 

The five clinging aggregates, in their assemblage, constitute sakkdya, 
the "existing body" or empirical personality. Therefore, on the grounds 
that things, i.e. personality and dukkha, equal to the same thing, i.e. the 
five clinging aggregates, are equal to each other, the structural formula 
of the four truths is occasionally stated in terms of sakkaya rather than 

dukkha. 1 Again, since all the five aggregates arise in connection with 
_____ ___ 

2. S. V. 12. 2. 3. Katamah ca bhikkhave dukkham ariyasaccaml 'Paiicupadana- 
kkhandha ti'ssa vacaniyamseyyathidanv.rupupadanakkhandho \edanupaddnakkhan- 
dho saHnupadanakkhandho sankharupadanakkhandho vinMnupadanakkhandho. 
Idam vuccati bhikkhave dukkham ariyasaccam. 

3. M. 44. 



TV. A. Jayawickrama 
Additional Abbreviations 

AA — Manorathapurani (Anguttara Nikaya Commentary: PTS, 5 vols., 

1924-57; reprinted 1964-73) 
Ap—Apadana (PTS, 2 vols., 1925-7) 

Dh A— Dhammapada Commentary (PTS, 5 vols., 1906-15; reprinted 1970) 
Divy. — Divyavadana (ed. E. B. Cowell and R. A. Neil, Cambridge, 1886) 
D.P.P.N. — G. P. Malalasekera, Dictionary of Pali Proper Names (India 

Office Library, London, 1937; PTS, 1960) 
I.A.— Indian Antiquary (Bombay, 1872-1933; 1964-) 
It.— Itivuttaka (PTS, 1890; reprinted 1975) 
J. A. — Journal Asiatique (Paris, 1822-) 
J.D.L. — Journal of the Department of Letters (Calcutta) 
J.P.T.S.— Journal of the Pali Text Society (London, 1882-1927) 
J.R.A.S. — Journal of the Royal Asiatic Society (London, 1843 — ) 
Katre — S. M. Katre, "Early Buddhist Ballads and their relation to the 

Older Upanishadic Literature" (Ph.D. thesis, London University, 1932) 
Kix.—Khuddakapatha (PTS, 1915; reprinted 1959) 
-Kvu—Kathavatthu (PTS, 2 vols., 1894-7) 
Lai. — Lalitavistara (ed. S. Lefmann, Halle, 1902/8) 
Mhv/Mvastu — Mahavastu (ed. E. Senart, 3 vols., Paris, 1882-97; tr. J.J. 

Jones, SBB, 3 vols., 1949-56) 
Nd.— Niddesa, Ciila (PTS, 1918); Malm (PTS, 2 vols., 1916-7) 
Reden — K. E. Neumann, Die Reden Gotamo Buddhos. Aus der Sammlung 

der Bruchstiicke. (Leipzig, 1911; Munich, 1924; Zurich and Vienna, 

S.H.B. — Simon Hewavitarne Bequest (Colombo) 
U.C.R.— University of Ceylon Review (1943-67) 
Vd.— Udana (PTS, 1885, reprinted 1948) 
Ud A— Udana Commentary (PTS, 1926) 
Vin.— Vinaya Pitaka (PTS, 5 vols., 1879-83; reprinted 1964) 
Vsm.—Visuddhimagga (PTS, 2 vols., 1920-1) 

Corrections to Vol. 1, No. 2: P. 75 — DA refers to the Sumangalavildsini 
(Dlgha Nikaya Commentary: PTS, 3 vols., 1886, 1931-2; reprinted 1968-71). 
P. 81, line 16 — Thai Shu refers to the Taisho (Tokyo) edition of the Chinese 


The Affhaka and the Parayana Vaggas appear to have been independent 
collections long before the existence of a separate work called the Sutta 
Nipata. The Culla Niddesa which comments on the Parayana Vagga 
and Khaggavisana Sutta and the Maha Niddesa which comments on the 
Afthaka Vagga form the eleventh book of the Khuddaka Nikaya. They 
make no specific reference to the Sutta Nipata. In spite of the fact that 
these two works were commentaries they came to be reckoned as canonical 
texts, and in turn were commented upon in the fashion of all canonical 
works. 1 The fact that the Atthaka and Parayana Vaggas and Khagga- 
visana Sutta had, at one stage, existed independent of a specific collection, 
does not necessarily prove that all other suttas in Sn. are late. The 
Niddesas themselves quote from suttas which came to be later included 
in Sn., besides quoting from other works in the Canon, and parts of Sn. 
already commented upon in the Niddesas. 2 


Asoka's Bhabrn Edict 

Some of the suttas included in Sn. are mentioned by Asoka in his 
Bhabru Edict (vide U.C.R. Yl. 2 p. 81), but often under different names. 
The Edict inculcates the study of the following passages:— 

1 . Vinaya-samukase, 

2. Aliya-vasani, 

3. Anagata-bhayani, 

4. Muni-gatha, 

5. Moneyya-Sute, 

6. Upatisa-pasine and 

7. Ldghulovade musdvadam adhigicya. 

1. Saddhammapajjotikd, the commentary on the Niddesas was composed during 
the reign of Aggabodhi I who ascended the throne (of Ceylon) in 554 A.C. (vide bctpj. 

' 2. Vide Nd 1 . ed. L. de la Vallie Poussin and E. J. Thomas pp. 513-515 and Nd 2 ed W. 
Stede pp. 289-290. Sabhiya Sutta is quoted from no less than 14 times, i.e. 
quoted at Ndl. 71, Nd 2. 220; Sn. 516 at Ndl. 244: Sn519 at Nd. 87 Nd 2 214; Sn 
522 (cp. A. III. 345) at Nd. 1.202, Nd2. 180; Sn. 527 at Ndl. 58, 221, 336; Sn. 529 atNd 
1. 93, 205, Nd 2. 256 and Sn. 531 at Nd 2. 255. Suciloma S. is Quoted from 4 times i.e. 
Sn. 271 at Ndl. 16, 364, 471 and Nd 2. 201; Pa dh ana S. also 4 "I^^L 7%79 Mrt 
at Nd 1. 96, 174, 333 and Nd 2. 253; Magandiya S. twice, viz. Sn. 844 at Ndl. 179,200, 
and Dhotakam&navapuccha (Sn. 1064), Mogharajamavavapuccha; (Sn 1119), Salla 
(Sn 576-581 ab cp. D. II. 120), Dvayatanupassana (Sn. 740-741) and Nalaka (Sn. 715) 
Suttas once each at Nd 1. 32, 438, 121, 455 and Nd 2. 118 respectively. 


Pali Buddhist Review 1, 3 (1976) 

Of these seven dhammapaliyayas (sections of the Scriptures) only Nos. 
1, 4, 5 and 6 have been observed by scholars to be identical with passages 
in Sn. All the seven passages are identified to some measure of satisfaction. 3 

Vinaya-samukase (1) "the Exalted Treatise on Moral Discipline" is 
identified with the Samukkamsika Dhammadesana (Ud. V. 3) by A. J. 
Edmunds in J.R.A.S. 1913 p. 387. Dr. B. M. Barua (J.R.A.S. 1915 p. 
809) identifies it with the Singalovada Suttanta (D. III. 180-194) arguing 
that Ariyassa vinaya which is the topic of discussion there is implied by the 
term Vinaya-samukase and that it was intended for the clergy and the laity 
alike. S.N. Mitra (I. A. 1919 pp. 8-11) suggests the Sappurisa Sutta (M. 
Ill, 37-45) on account of the occurrence of the words vinayadhara and 
attanam samukkamseti. Bhandarkar (Asoka pp. 87-88) attempts to prove 
its identity with the Tuvataka Sutta of Sn. (Sn. 91 5-934) from the fact that 
it is included by Buddhaghosa in a list of four suttas, three of which can be 
identified with three of Asoka 's dhammapaliyayas. He adduces further 
interval evidence and maintains that the Buddha expounds religious 
practices here, for, pafimokkha, patipada and samadhi are some of the 
topics under discussion. 

Muni-gStha (4) is undoubtedly the : . Muni Sutta o?Sn.(Sn. (Sn. 207-221). 4 
Rhys Davids (J.P.T.S. 1896 p. 95) argues that if Saila-gatha (at Divy. 35) 
meant Sela Sutta, then Muni-gatha should be the Muni Sutta. He further 
states ' 'that Asoka should lay so much stress on this short poem is only in 
harmony with the tenor of the whole context in the Edict". 

The next dhammapaliyaya Moneyya-sute (5), is identified with the 
discourse of the Nalaka Sutta (Sn. 699-723). It was wrongly identified as 
either A.I. 273 or It. 56 (Rhys Davids loc. cit.); but all available evidence 
shows that Moneyya-sute was none other than the Na/afta-discourse. The 
alternative name for the Nalaka Sutta in Pali itself is Moneyya Sutta 
(Chalmers xi), which perhaps owes its origin to the opening word money- 
yam. Further, the Sutra in Mvastu. that corresponds to this discourse is 
also called Mauneya (Mvastu. III. 387 ff.). The short and unimpressive 
prose passages at A.I. 273 and It. 56 could not in any way have been the 
Moneyya-sute of Asoka, though they deal with Moneyydni in brief. 

Oldenberg and Rhys Davids attempt to identify Upatisa-pasine (6) 
with a Vinaya passage {Vinaya Texts 3. 149 i.e. Vin.I. 39-41) which gives 
the story of Sariputta's conversion as a result of his question to Assaji. 

3. Vide Hultzsch, Corpus Inscriptionum Indicarum, Vol. I, pp. 172-174 ff.; Dharmlk- 
nanda Kosambi, I. A., 1912 pp. 37 ff; Rhys Davids, J.P.T.S. 1896 pp. 93 ff; J.R.A.S. 
1898 pp. 683 ff; Radhakumud Mookerji, Asoka, pp. 117 ff; D. R. Bhandarkar, Asoka, 
pp. 85 ff; J.D.L. (Calcutta) xx, pp. 1-7; Sylvain Uvi, J. A. 7, 475 ff; and Oldenberg, 
Vinaya Pifakam I, xl ff. 

4. Vide Dharmtinanda Kosambi, 1. A. 1912 pp. 37 ff; Mookerji, Asoka pp. 16 ff. 
and Bhandarkar, Asoka 85 ff. 


Pali Buddhist Review 1, 3 (1976) 


Rhys Davids elaborates furtheron this inJ.R.A.S. 1893 p. 693 and J.P.T.S. 
1896 pp. 97-98. But Dharmananda Kosambi (I.A. 1912 p. 40) identifies it 
with Sariputta Sutta (Sn. 955-975). It is generally accepted that the pass- 
ages mentioned by Asoka are short pieces. The people were instructed 
to study these dharmaparyayas and perhaps learn them by heart as was 
the practice then. A passage in verse lends itself easier for memorising 
than one in prose, and has more poetic appeal. This alone is sufficient 
reason why Upatisa-pasine cannot be the prose sutta at Vin. I. 39-41. 

The seventh "section of the scriptures" called the "Exhortation to 

Rahula, beginning on the subject of Falsehood" has so far been identified 

as the Ambalatthika Rahulovada Sutta (M.I. 41 4-420) , 5 but the probability 

is that it perhaps referred to a Rahula Sutta in verse. The only Rahula 

Sutta in verse in the Pali Canon, is found at Sn. 335-342. But the sutta 

as it exists now, cannot be easily identified with Laghulovade musavadam 

adhigicya, as it neither begins with (adhi+^kr), nor deals with the topic 

oimusavada (falsehood) anywhere in the body of the sutta. It has been 

pointed out by Katre that probably the Vatthugatha (Sn. 335-336) formed 

a part of a different Rahula Sutta and that the concluding sentence in prose 

links them with the rest of the sutta. He further states that the clue to 

the verses is found only in the prose formula at the end of the sutta. This 

other Rahula Sutta, presumably a part of which is now preserved as 

Vatthu-gatha in Sn. was probably the sutta mentioned by Asoka. But 

all this is purely conjectural. No definite connection can be established 

between Sn. 335-336 and MI. 414-420, the other Rahula Sutta; and there 

is no conclusive proof that No. 7 in the Edict had any connections with 

Sn. 335-336 or Sn. 335-342. The only reasons for suspecting that they 

were connected are: — 

1. The Rahula Sutta in Sn. is a comparatively short piece in verse. 

2. The two, Laghulovada and Rahula Sutta refer to the same person 


3. This dhammapaliyaya follows three others in the Edict which are 

identified with certainty to belong to the same type of literature 

(i.e. pieces now preserved in Sn.). 
Eliminating the Rahula Sutta as doubtful there yet remain four suttas of 
Sn. in Asoka's list. The consensus of opinion among scholars is that 
Munigatha, Moneyya-sute and Upatisa-pasine referred to suttas which 
were included in Sn. Perhaps Bhandarkar is correct when he identifies 
Vinayasamukase as the Tuvafaka Sutta. 6 There is no doubt that these 
suttas existed at least as early as the 3rd century B.C. For lack of further 
evidence it is incorrect to presuppose the existence of Sn. prior to the time 

Rhys Davids J.P.T.S. 1896 p. 95. 
op. cit. 


Pali Buddhist Review 1, 3 (1916) 

of Asoka as there is no specific mention of it either in inscriptions or in any 
Canonical work. 


The Atthaka Vagga 

The Episode of Sona Kutikanna. 

On the other hand, the early existence of the A((haka and the Pdrdvana 
Vaggas as separate collections, can be deduced from the references made 
to them in other works. The earliest mention of the Atthaka Vagga is at 
Vin. I. 196, in the episode of Sona Kutikanna, which repeats itself in many 
other works with various additions and alterations. 7 The Vinaya passage 
runs . . ayasma Sono sabbdn'eva Atthakavaggikani sarena abhdsi (the 
venerable Sona recited all the sections — or suttas — of the Atfhaka Vagga 
with proper intonation). At Ud. 59 the precise number of suttas in the 
Atthaka Vagga is also mentioned . . ayasma Sono . . solasa Atthakavaggikani 
sabbdn'eva sarena abhani (the venerable Sona recited all the 16 — suttas — 
of the Atthaka Vagga with proper intonation). Dh A. IV. 102, UdA. 312, 
AAA. 241 and Th 1 A. I. 459 relate this incident in very much the same 
words, but with additional commentarial gloss. 

The Avadana of Ko|ikarna (Divy. 20), which is an extract from the 
Vinaya of the Mulasarvastivadins 8 mentions the Atthaka Vagga: Athdyus- 
mdnchrono bhagavata krtavakasah asmdtpardntikayd guptikaya Uddnat, 
Pdrdyandt, Satyadrstah, Sailagdthd, Munigdthd, Arthavargiyani (v.i. 
arthavadgiydni) ca sutrdni vistarena svarena svddhydyam karoti. (Then 
the venerable Srona, with the approval of the Bhagava, rehearsedin detail, 
with intonation in the accent of an Aparantika, passages from the Uddna 
and Pardyana, the Satyadrsta (?), the Saila-gathd (Sela S), Munigdthd 
(Muni S.) and the sutras of the Arthavarga). 

In the Avadana of Purna at Divy. 34-35, the merchants who embaTked 
with Purna are said to have recited the Uddna, Pardyana, Satyadarsa, 
Sthaviragdthd, Sailagdthd, Munigdtha and the Arthavargiya Sutra. 

In the Vinaya of the Sarvastivadins which is found in Chinese 9 (Tok. 
XVI. 4. 56a), Srona is said to have recited the Pardyana and the Satyadaria. 
The Buddha complements his AvantI pronunciation. 

7. Tlie episode of Sova (Srona) in Pali andBSk. is fully analysed and critically studied 
by Sylvain Uvi in J. A. 1915 pp. 401 ff. 

8. Vide Huber, B.E.F.E.O. 1907, Sylvain Uvi, T'oung Pao 1907 and M. Choyannes, 
Cinq cents Comes et- Apologues II, 237 ff. . 

9. I am indebted to Prof. Sylvain Levi's analysis of the Srona Episode in J. A. 1915, 
for these references. 

Pali Buddhist Review 1, 3 (1976) 141 

The Vinaya of the Mahlsasakas, preserved in Chinese (Tok. XVI, 2. 30a), 
contains a version similar to the Pali account in the Vinaya; but the number 
of suttas is specified as in the Uddna. 

The account in the Vinaya of the Dharmaguptakas resembles the version 
in Pali and the account in the Vinaya of the Mahlsasakas. Here (Tok. 
XV. 5. 53b; chap. 39) Kotikarna is said to have recited the 16 Arthapada 
without addition or omission. 

In the Vinaya of the Mahasanghikas (Tok. XV. 9.61a; chap. 23) Srona 
recites the Astavarga (Ch. Chu Pa-cWun-ching), and the Buddha questions 
him on the phrases (padd) and the meaning (arthd). 

In all these accounts, except in the Vinaya of the Sarvastivadins, the 
Arthavargiyani or the Atthakavaggikani are mentioned. The additional 
list of titles in Sanskrit texts is a mere expansion though Sylvain Levi does 
not consider it an interpolation: 

' 'On pourrait etre tente de croire que la liste des titres donnee dans le 
texte Sanscrit est une interpolation, si la version tibetaine du Dulva ne 
venait pas controler — et sur certains points rectifier — le texte Sanscrit. ' ' 
(ibid. p. 412). The Tibetan version parallel to the Vinaya of the Mulasa- 
rvastivadins is at Dulva I. 378-405 (cf. Divy. I, Ch. Tok. XVII. 4. I04e- 


Other References in Buddhist Sanskrit Works 

Besides the episode of Sona (Srona), there are numerous references to, und 
quotations from the Atthaka Vagga. Vasubandhu in his Abhidharmakoka 
quotes the following verse: — 

Tasya cet kamaydnasya chandajdtasya dehinah 

te kdmd na samrdhyanti salyaviddha iva rupyate 
and attributes it to the Atthaka Vagga (cp. Sn. 767). 

Yasomitra in his Abhidharmakosa-vydkhya comments: Tathd hyartha-. 
vargiyesuktam iti: Arthavargiyani sutrdni ydni Ksudrake pathyante tesvk- 
tam; tasya cet kamaydnasya iti vistarah. Tasya dehinah kamaydnasya 
chandajdtasya yadi kdmd visaya na samrdhyanti na sampadyantc salyaviddha 
ivdsau rupyate bddhyate ityarthah. (This is what is meant by the state- 
ment that it has been so said in the Arthavargiyas: Jt is stated in the 
sutras of the Arthavargiya found in the Ksudraka (Nikdya or Agamal) that 
the meaning of ' 'if of him who desires etc. " is, ' 'If the desires and sense- 
pleasures of a being who yearns and craves for such pleasures are not 
satisfied nor fulfilled he sulks and is perturbed like one shot with an 


Pali Buddhist Review J, 3 (1976) 

Bodhisattvabhumi (p. 48) commenting on the word kanti says thus, 
Vktam ca bhagavatd Arthavargiyesu, 

Yd kascana samvrtayo hi loke, sarva hi td munir nopaiti 

Anupago hyasau kena upddafita, drstasrute kantim asamprakurvan 

(cf. Sn. 897). ' 'And so it has been said by the Bhagavain the Arthavargiyas, 
'Whatever conventions of the world there are, none of them affects the 
muni (sage), for he does not move with them, wherefore shall h? who forms 
no sense-attachment to what is seen and heard be guided (by them)?' " 

The reading kantim in Bodhisattvabhumi sheds a new light on the inter- 
pretation of the line Sn. 897 d. All MSS., except two Burmese MSS. 
(Nos. 4 and 5 mentioned at Sn. p. v., P.T.S. ed.) which read khanti, agree 
on the reading khantim. Nd 2. 165 considers khanti as a synonym for 
ditthi, ruci, laddhi, ajjhasaya and adhippdya, perhaps guided by the 
occurrence of dittha and suta at other passages in Sn. 10 Sn. A. 558 
comments on it as: khantitnakubbamdno, ti. pemam akaronto. Chalmers 
translates Sn. 897d as, "when phenomena of sense appeal to them no 
more," Fausboll, "he who is not pleased with what has been seen and 
heard," Neumann, Beitn Sehn und Horen angehalten nimmer, and E. M. 
Hare, ' 'why give accord to things of sight and ear?" All these translators 
apparently translate the idea correctly, but none Of them seems to have 
questioned the text. According to the reading khantim the corresponding 
Sk. would be ksantim (patience). 11 The word khanti in a context like this 
may be translated as, ' 'tolerance for ' ' or even ' '(developing) a weakness 
for/ ' but such a translation appears unnatural and laboured. If the text 
had been kantim (fromv^am) and not khantim, the idea conveyed would 
be more in keeping with the context. The reading khantim in BSk. 
cannot be brushed aside as a wrong Sanskritisation for Pali khantim. 
On the other hand, it may perhaps go back to a reading much earlier 
than Pali. 

There are also a number of passages and lines common to the Atthaka 
Vagga and other Pali works. They are fully examined by Franke. 12 

The Chinese Arthapadam 

The complete Atthaka Vagga together with additional stories as a back- 
ground is found in the Chinese Tripitaka although "it can be said with 

10. This idea occurs no less than 18 times in Sn. viz. Sn. 797 b =887 a, 793 ab =914 ab, 
798 cd, 887 ab (887 b = 790 b), 910 ab, 1079 ab = 1080 be, 1082 cd=1083 cd. 788 b = 
789 a, 802 ab, 897 d and778 d=250 d. All these instances (except 250 d) are, found inthe 
sections commented in Nd 1 and 2. 

11. Vide E. J. Thomas, History of Buddhist Thought, p. 171 and s.v. P.T.S. 

12. Vide R. Otto Franke, Die Sutta Nipata Gathas mit ihren Parallelen, Z.D.M.G. 
1909-1912 andE. M. Hare, Woven Cadences, (S.B.B. Vol. XV),pp. 203-206. 

Pali Buddhist Review 1, 3 (1976) 


certainty that there is not and has never existed a Chinese version of the 
Sutta Nipata. ' ' 13 This section called the 1-tsu or Yi-tsou king (Arthapadam) 
is a translation dating back to the beginning of the 3rd century A.C., 
according to Anesaki. It occurs as No. 198 in the Taisho Tripitaka. The 
16 pieces occur in the following orden- 

Kama Sutta contains 

GuhafihakaS. „ 16 

Duflhatfhaka S. „ 4 & 12 

Suddhaffhaka S. „ 16 

Paramatthaka S. ,. 16 

Jar a S. 20 

Tissametteyya S. ,. 20 

Pasura S. „ 23 

Magandiya S. ,, 27 

Kalahavivdda S. ,, 32 

Culaviynha S. ,, 34 

Mahdviyiiha S. „ 40 

Tuvafaka S. „ 40 

Sariputta S. ) 6 & 24 

Purabheda S. „ 28 

Attadanda S. ,, 40 

8 lines with 3 padas each 
















(No. 12 
(No. 13 
(No. 14 
(No. 16 
(No. 10 
(No. 15 

1 in 


Pali Atthaka Van 




In addition to the prose incorporated with these verses there occur some 
additional stanzas. The lines beginning with na socanaya at A. II. 62 
are found at No. 1 and Sn. 152-179 in No. 13 of the Chinese version, i.e. 
Tuvataka Sutta. 

All this evidence helps to show that the Atthaka Vagga as a collection is 
old, and Sylvain Lovi 1 - 4 concludes, Nous sommes en droit de classer V Ar- 
ihavarga parmi les monuments les plus auciens de la literature bouddhique. 

Astaka Varga or Arthaka Varga? 

The title A({haka Vagga calls for attention next. The name Atthaka 
suggests that the- vagga consists of octaves or suttas with eight stanzas 
each, but only four of its suttas (viz. Nos. 2-5) are proper octaves. It 
cannot be determined whether these suttas were atthakas(astakas— octaves) 
or atthakas (arthakas — meaningful utterances) to begin with. Pali 
tradition has been very strong in insisting on the name Atthaka. It was 
customary for Pali compilers to resort to artificial means such as numerals, 
in their. classifications. They may have deemed it proper to call a section 
Atthaka Vagga even though only a small proportion of its suttas consisted 
or real octaves. Similar instances may be seen in works like udana where 
an important sutta in a vagga gives the name to the whole of it. It was 
not considered necessary that all the suttas in the vagga should consist of 
8 stanzas each, unlike the majority of the nipatas (the earlier ones) of 


Anesaki, J.P.T.S. 1906-1907, p. 50. 
J. A. 1915, p. 417. 


Pali Buddhist Review 1, 3 (1976) 

Th 1 and Th 2. It would be incorrect to say that only these four atthakas 
formed the Atthaka Vagga and that the other suttas were subsequently 
added or grafted from other places. This would imply a tacit acceptance 
of the incomplete artificial classification of Pali compilers as final. In 
fact, the vagga follows a systematic arrangement in which the sutta with 
the least number of verses is placed first and proceeds gradually in ascend- 
ing numerical order till the suttas with the highest number of verses are 
placed last. The order of arrangement of the suttas need not necessarily 
be as old as the vagga itself, for the Chinese version follows a slightly 
different order. However, nothing conclusive can be inferred from this. 

Almost all the references to the Atthaka Vagga which mention the 
number of suttas in it speak of the Solasa atthakavaggikani (Ud. 59, Ud A. 
312, DhA. IV. 102 and AA. I. 241). The Chinese version was seen to 
contain the 16 suttas in full. Despite this general agreement ThIA. speaks 
of ' 'addhuddhasolasa atthakavaggikanV " ': (ThIA. I. 459 S.H.B., commen- 
tary on Sonatthera's verses at Th 1. 365-369). This statement would 
imply that the Atthaka Vagga consisted of 56 (3^x16) suttas — an im- 
possibility. Commentarial tradition cannot always be relied upon; and 
in all probability this statement may have been an exaggeration like the 
passage at AA. IV. 35 which speaks of 250 stanzas of the Pardyana, when 
in actual practice the whole vagga, including the Vatthu-gathas and Epilo- 
gue contains only 174 stanzas. The statement at Th. 1 A.I. 459 can also 
be interpreted as "56 stanzas of the Atthaka Vagga." It is not possible 
to find out to what suttas these 56 stanzas, belonged. Obviously the 32 
stanzas which form the four regular octaves should be included in this 
number. This leaves 24 verses which should be expected to belong to 
three other regular octaves; but no such suttas are to be found in the vagga. 
Furthermore, there is no possible combination of two or more suttas 
which brings about a total of 24 stanzas. There is no justifiable reason 
why a commentary of so late a date as 5th century A.C. should ignore 
some of the suttas and speak of only 56 stanzas when Nd 2. Vin. I. 196 
and Ud. 59, leaving aside contemporary commentarial literature, confirm 
that it did consist of 16 suttas. The reading, addhuddhaso[asa atthaka- 
vaggikani is incompatible with evidence furnished by all other sources 
and therefore can be summarily dismissed as a Commentarial error. 

Pali works uniformly refer to this section as Atthaka Vagga though BSk. 
and Chinese Buddhist works give it different names. It is called Arthavar- 
giydni Sutrani (v.l. arthavadgiydni) at Divy. 20 and 35. Vasubandhu and 
Yasomitra (supra) call it Arthavargiya. Bodhisattvabhumi too refers to it 
as Arthavargiya. The Chinese version gives the name as I-tsuo or Yi-tsou- 
kmg (Arthapadam). In the episode of Srona found in the Vinaya of the 

Pali Buddhist Review 1, 3 (1976) 


Mahlsasakas 15 (Ch. Tok. XVI. 2. 30a) the reference is to the sixteen 
Arthakavarglya (Ch. Yi-pin= Artha-varga). The Vinaya of the Dharma- 
guptakas (Tok. XV. 5. 53b) has it as the sixteen Arthapada (Ch. Yi-kiu= 
Artha-pada). In the Vinaya of the Mulasarvastivadins it is called the 
Arthavarglyani Sutrani (Dulva: don kyi choms kyi mdo). The Sarvastivad 
in Vinaya calls it the Artha-varglya Sutra (Tok. XVII. 4. 9b, Col. 5; Ch. 
Yi-tsing=Artha-varglyd). At Tok. XVI. 4. 56a it is called A : tch'a-p' 
o-k'i-ye-sieou-to-lou, sutra des vertues rassemblees and is identified by Levi 
(ibid.) as Arthavargiya Sutra. 

It is significant that the majority of these works refer to it as Artha (ka)- 
varga or Artha-pada. The Vinaya of the Mahasanghikas alone speaks of an 
Asta-varga, but even here the idea of artha and pada is not absent. It is 
said that .the Buddha questioned Srona on the phrases (pada) and the 
meaning (artha) after his recital (Tok. XV. 9. 61a, chap. 23). The Vinaya 
of the Mahasanghikas thus preserves the Pali tradition at the same time 
reflecting another common to the rest of the BSk. works. It is quite 
probable that this section was originally not meant to be described as the 
"Eights," and BSk. may have preserved an earlier tradition which called 
these Atthakas Arthakas. The four octaves were probably Arthakas 
(Atthaka-mzaxim$xx\ statements) at the beginning. Each of these suttas 
contains in its opening line the words used for their respective titles. 
Guhaffhaka opens with, Satto guhayam bahunabhicchanno (Sn. 772a), 
Dutthatfhaka with, Vadanti ve dutthamana, pi eke (Sn t 780a), Suddafthaka 
with, Passami suddham paramam arogam (Sn. 788a) and Parama((haka 
with, Paramam ti ditthisu paribbasano (Sn 796a). These words are used 
as illustrations in the didactic-ballad discourses to elucidate the meaning, 
and hence the suttas are Atthakas (Arthakas). It is a mere coincidence 
that the number of stanzas constituting each of these suttas happens to 
be eight. The word attha together with the secondary suffix — ka (attha+ 
ka) may have changed into atthaka (probably) with the influence of Western 
Prakrit which has a tendency to cerebralise dentals following an r; i.e. — 
rt— > — tt— and— rth— > —tth—. The first change is frequent in Pali 
itself; e.g. Sk. arta > at(a. Artha itself is frequently changed to attha, 
which spelling was later restricted to a specific meaning as "law-suit" 
(a(ta). In cpds. artha > attha in Pali e.g. at\hakatha, atthuppatti, etc (s.v., 
P.T.S.). In the case of the Atthaka Vagga this change perhaps was more 
accentuated by the mere coincidence that four of its suttas consisted of 8 
stanzas each. The weakness of Pali compilers to be guided by numerical 
classifications may have finally led to stamp down the name "eights" or 
"octaves" on this vagga. 
The emphasis on attha (weal) in the Pali Canon is evident from the 
15. The following references to Chinese works are from Sylvain Levi, ibid. 


Pali Buddhist Review I, 3 {1976) 

numerous instances in which the word occurs. 16 The formula, 
atthaya hitdya sukhdya (for the benefit, well-being and comfort of) which 
occurs all over the Canon (e.g. D. III. 211 ff. It. 79, Kh. VIII. 1 etc.), 
leaving aside all other references to attha, testifies to the importance of 
this concept. It is probable that the idea underlying the Atthakas of the 
Atthaka Vagga was related to attha (weal) though fundamentally it was 
the elucidation of meaning {attha paridipana) that was aimed at. This 
may be seen more clearly at Saddharmapundarlka 383 1.3. 

Evam idam mahdrthasya dharmaparydyasya dhdrana, vacana, desand 
Bodhisattvdnam anuttaraydh samyak sambodher dhdraka samvartanti, 
(In like manner, the learning by heart, the reciting and the teaching of this 
section of the scriptures of great meaning — or benefit — -tends to bring 
about the perfect and supreme Enlightenment of Bodhisattvas. 

All this evidence seems to indicate that the term Atthaka Vagga (also 
inahdsanghika Asta-varga) was a, misnomer arising from an early confusion 
caused by the occurrence of eight stanzas each in Nos. 2-5 of the Vagga. 
The term Atthaka is best interpreted as Arthaka as in the majority of 
BSk. works. 

From isolated references to Atthaka Vagga in many Pali works it is 
conjectured that "it may possibly have been the name of divisions of 
other works. ' ' 17 In the whole of the Pali Canon no other Atthaka Vagga 
can be traced though Anguttara has an Atthaka Nipdta and Th 1 and Th 2 
contain Attha Nipdtas. Though the absence of other Atthaka Vaggas 
does not preclude the possibility of the occurrence of other sections 
bearing that name no references to another Atthaka Vagga have been 
discovered so far. 


Parayana Vagga 

The next vagga in importance is the Parayana. It consists of 18 pieces; 
viz. a prologue in verse called the Vatthu-gathd, 16 short dialogues in verse 
called Pucchas and an epilogue in prose and verse. The word Parayana 
occurs thrice in the text itself, but all these references are to be met with in 
the epilogue; viz. Sn. p. 218, Sn. 1130d and 1131a. The prose passage at 
p. 218 gives a commentary-like explanation of the term Parayana: Ekam 
ekassa ce pi pahhassa attham anndya, dhammam anndya, dhammanudham- 
mam patipajjeyya, gaccheyy' eva jaramaranassa param, paramgamaniyd 
/'me dhamma' ti, tasmd imassa dhammapariyayassa pdrayanam Vveva 

16. Vide P.T.S. (s.v.) for examples quoted. 

17. Matalasekera s.v., D. P.P.N. 

Pali Buddhist Review 1. 3 {1976) 147 

adhivacatiam. (If one were to comprehend the import of each one of these 
questions, and realise the Dhamma therein, and follow the path in accord- 
ance with the major and minor precepts of the Law, one would cross over 
to the further shore of old-age and death. As these teachings lead to over- 
yonder, the name Parayana is given to this disquisition on the Dhamma). 
The two stanzas Sn. 1 129-1 130 express the same idea in verse and explain 
the title Parayana. 

Although the title does not occur in any of the Pucchas (or panhas) the 
central theme of the vagga is "The Way Beyond" or "Crossing Over." 
The idea of crossing over of the Flood (pgha) occurs 10 times. 18 The 
"passing beyond" of this "Sinful State" {visattikd) is mentioned 5 
times, 19 and this is an idea common with other canonical texts, particularly 
Samyutta and Anguttara Nikdyas. The overcoming of birth and old-age 
{jati and jard) which is a necessary accompaniment of the ' 'Going Beyond 
is to be met with in 10 places. 20 An idea parallel to this is the abandoning of 
(y/haot pa +Vhd) sorrow, or that of jdtijara (or jati and jam), occurring 7 
times in Sn. 21 Connected is the idea of overcoming the material substratum 
of birth, (ttpadfti) at S«- 1057b and 1083b. The destruction ot(pa+Vbhid) 
ignorance (avijjd) occurs at Sn. 1105 f and 1078d, and of craving {tanhd) 
and attachment {kdma and its synonyms), 9 times. 22 The other concepts 
emphasised are, the state of emancipation {vimokha) at Sn. 1088d, 1105e, 
and 1189d (the Buddha is called vimutta, the released, at 1101 and the 
emancipated one is mentioned at Sn. 1071c, 1072c, 1073c, 1074c and 
1114d) cessation {nirodha) at Sn. 1037e, the destruction {uparodha, or 
verb upa+y/rudh) of evil at Sn. 1036e, 1037df, 1110b, lllld, tranquility 
{santi) at Sn. 1066a, 1067a, the tranquilled state {santipada) at Sn. 1096c, 
nibbana at Sn. 1061 d, 1062d, 1094c, 1108d, U09d and nibbanapada at Sn. 
1086d. Ajita questions the Buddha regarding the taints of the world at Sn. 
1032; the dangers arising out of the world are mentioned at Sn. 1032, 1033, 
ofogLat 1092, l093andofsorrowandthearisingofIllatSn. 1033, 1049, 
1050 and 105 1 . The escape from the evils of the world, the crossing over 
of the Flood and the attainment of santi or nibbana are the dominant ideas 
in the vagga. The verb mth\/tar alone is used no less than 23 times in 
the Pucchas in addition to verbs like pajahati, thus justifying the title 

The word para occurs thrice in the Pucchas {Sn. 1059, 1105 and 1 1 12); 
but in the latter two instances it is used in praise of the Buddha. In the 

IS Sn 1052 c, 1059 c, 1064 d, 1069 d, 1070 b, 1081 e, 1082 g, 1083 g, 1096 b and 1101 b. 

1 £ S 'd%iS X Af^,f?A TuA"t"Si. «. no, - ,02, 

il is used as an epithet of the Buddha). 


Pali Buddhist Review J, 3' (1976) 

whole of 5/7. the word occurs 43 times, together with its derivatives and 
cpds., evenly distributed in all the five vaggas. Of these, para, "the 
Beyond," is directly mentioned in five instances; viz. na param digunam 
yantiiSn. 714c), tinno ca param akhilo akankho K (Sn. 1059d), gacche param 
aparato (Sn. 1129d) and maccudheyyaparam (Sn. I146d). 'The idea of 
"crossing over" is incorporated in a simiie at Sn. 771d, and parasmim 
(he.) occurs at Sn. 1018c and 1020d. This concept is totally different 
from parami or paramita of later Buddhism. Paramgata occurs at Sn 
803d and paragata at Sn. 21b, 210d, 359b and 638c. Para in the line, so 
bhikkhu jahati oraparam (Sn. lc-17c— that monk shuns the here and the 
beyond) has a different connotation horn para in the rest of the references. 
The idea that is diametrically opposed to param+Vgam is at Sn. 15 b, 
oram agamanaya paccayase (casual antecedents for the return hither).' 
The concept of "going beyond" is to be met with in numerous other 
canonical works; e.g. S. IV. 174, A.V. 4, M. III. 64, Th 1. 771-773, etc. 
and is one of the most fundamental tenets in early Buddhism. 


Its Antiquity 

This vagga appears to have been called Parayana from the earliest times. 23 
Several canonical works refer to it and quote from it. Sn. 1 109 is found 
at S. I. 39 in the Devata Samyutta, and at S. I. 40 the same stanza occurs 
with its first line reading, nandi sambandhano loko instead of nandlsamyo- 
jano loko. Yet there is no mention of the Parayana here. S. II. 47 refers 
to the Ajitapanha when quoting Sn. 1038, and the stanza is quoted again 
at 5". II. 49 making it the topic of discourse up to p. 50. Anguttara refers 
to the Parayana 6 times. At A. I. 133 Punnakapahha of the Para- 
yana is mentioned and Sn. 1048 quoted. At A. II. 45-46 the same stanza 
is quoted thus: Ima kho bhikkhave catasso samadhibhavana, idam panaetam 
sandhaya bhasitam Parayane Punnakapanhe (These indeed, O monks, are 
the four meditations on concentration; it has been declared so in the 
Punnakapafiha of the Parayana regarding this). The Udayapanha of 
the Parayana is mentioned at A. 1 134, and Sn. 1 106, 1 107 are quoted from 
it. A. III. 399, 401 quoted. 1042 with the opening line reading differently 24 
and refer to the Metteyyapahha of the Parayana. At A . IV. 63 the female 
lay-devotee Nandamata is reported as reciting the Parayana with proper 
intonation (sarena) and Vessavana is pleased with it. Sn. 1064 is quoted 
atJfrM. 94; Sn. 1117 at Ap. 537, 25; Sn. 1 118-1119 at Ap. 537, 26-28; 

23. Also Vide\ 111; Anesaki J.P.T.S. 1906-7 p.Tl^mentions that no less than 13 
references are made to it in early texts. 

24. Sn. 1042a reads, so ubhantam abhinnaya, while the line at A III 399 readi vn 
ubhante vuhtvana. ' '' 

Pali Buddhist Review J, 3 (1976) 


Sn. 1119 at Vsm. 656 and Kvu. 64; besides the numerous instances where 
the verses of the Parayana are quoted in Commentaries and quoted 
and commented on in Nettippakarana. 25 

Among the references in BSk. works 26 many of the following have 
already been mentioned in connection with the Atthaka Vagga. Divy. 
20, 34, include it among the texts recited by Srona and the merchants 
respectively. The Dulva too mentions it in the episode of Srona. The 
Vinaya of the Sarvastivadihs (Ch. Tok. XVI. 4. 56a) mentions Po-lo-yen 
(Parayana) as one of the texts recited by him, and includes it among the 
"great sulfas" in a passage which is important for the history of the 
Canon. Po-lo-yen (The Way Across) is the 16th passage out of the 18 
mentioned. The Arthavargiyasutra is No. 17, and the majority of the 
other passages is from D. In a list of sutras "which should be taught to 
novices" occurring in the Vinaya of the Mahasanghikas (Ch. Tok. XV. 
8. 9. 3a) the Pa-ch'ung-ching (Atthaka Vagga) and Po-lo-yen are mentioned 
at the head. The Vinaya of the Dharmaguptakas (chap. 54) too refers to 
the Parayana. Reference to it is also made in Abhidharmamahavibhasa 
(chap. 4), where it is stated that the Parayana was recited at a ' 'Council" 
of 500 arhats held under Kaniska's patronage. Some of the passages 
specially cited are the second stanza of Posalamanavapuccha and Sn. 874. 
Mahdprajhaparamita Sastra in its first chapter quotes the "Question of 
Makandika" in the Atthaka Vagga (Sn. 837-840 are quoted), and in 
chapter 3 ' 'The Question of Ajita in the Parayana 1 ' (Sn. 1032 ff). Asvag- 
hosa refers to the brahmanas of the Parayana in his Buddhacarita (v. 1061) 
and Sutralankara (canto 43). E. J. Thomas (Life of Buddha, p. 274) 
mentions the story of Bavari in a later form found among the MSS. dis- 
covered in Central Asia and cites Sieg und Siegling, ' 'Tocharische Sprac- 
hreste" I, p. 101. 

From all these references, specially those in the Pali Canon, which are 
older than the BSk. works, it is evident that the Parayana existed very early 
as a separate collection. Nowhere is Sn. mentioned when quotations 
are made from various panhas. This is further proof that the Parayana, 
like the Atthaka Vagga goes back to a period prior to the compilation of 
Sn. The various quotations also show that the questions of the Parayana 
have not undergone the rigid classification and arrangement found later 
in Sn. They are invariably called panhas and not pucchas unlike in Sn. 

W. Stede (Nd 2, p. xx) suggests that these panhas may have existed 
in ' 'some arrangement other than that which enumerates them simply as 
Pucchas 1, 2, etc." By carefully analysing the various MSS. of Nd 2 he 
notes that the Niddesa makes it ' 'conclusive to a certain extent that groups 

25. Vide Otto Franke and EM. Hare, ibid. 

26. Op. cit. (Sylvain LSvi). 


Pali Buddhist Review I, 3 (1976) 

of pucchas existed separately before they were set in the present arrange- 
ment, or were taken out of their present setting because they were greater 
favourites than others." The popularity of the Ajita Sutta perhaps led 
to its being placed at the head of the vagga. Metteyya and Punnaka 
Panhas can be considered to have been equally popular, judging from the 
quotations made from them in Pali works; and this probably explains 
their position as second and third respectively in the vagga. Udaya 
Pahha is also quoted from, but it is placed as No. 13. Stede concludes 
that either of Nos. 3 and 4 may have formed the last sutta of a separate 

It is generally accepted that Nd 2 is older than Sn. The latter does not 
yield any information regarding the arrangement of these pucchas. All 
the 16 pieces are called Pucchas, whereas in Nd 2 some are called suttas; 
(viz. Nos. 1 and 3, and the others are called panhas). The minor variations 
in the mode of referring to and commenting on these pieces in Nd 2 may 
shed some light on this question. Stede 27 shows that Nd 2 is uniform as 
regards the concluding statements in the Commentaries of the suttas up 
to No. 3; e.g. Ajita Sutta Niddeso samatto, etc. that Nos. 4 and 5 are 
numbered after the comments on them (e.g. Mettagu pahham catuttham 
samattam, etc.), and that the numeration ceases after No. 5. He questions 
whether Nos. 1-5 formed one separate collection. It is quite probable 
that Nos. 1-3 formed one collection and that Nos. 1-5 another, so that the 
group Nos. 1-3 was either included in the bigger group Nos. 1-5, or the 
earlier group was Nos. 1-3 which was later extended up to No. 5. It is 
quite obvious that Nos. 6-16 formed a group or groups independent of 
Nos. 1-5. The position of the popular Udaya Panha as No. 13 may 
suggest that it may have been placed at the head of another group consist- 
ing of Nos. 13-16 just as the well-known Ajita Panha was placed at the 
head of the earlier group (Nos. 1-3 or 1-5). The probability is that Nos. 
6-16 consisted of two groups viz. Nos. 6-12 and 13-16. All these pieces 
were, at a subsequent date, taken together and gradually worked out into 
a legend by introducing Baravl, the brahmin of the South. 


The Vatthu GStha 

The legend of Bavarl leads to the question of the relationship oj the vatthu- 
gatha and the epilogue to the pucchas of the Par ay ana. The Niddesa leaves 
the vatthu-gathd (v.g) uncommented and it is doubtful whether they were 
known to its author. In some MSS. of Nd 2 (vide Nd 2 introduction) 

Pali Buddhist Review 1, 3 (1976) 


the text of the v.g. and that of the epilogue is to be met with, while in others 
only that of v.g. The inclusion of the v.g. and the epilogue in MSS. of 
Nd 2 does not help one to determine whether the author was acquainted 
with these two pieces, for it may have taken place long after the writing 
of Nd 2. The fact that the v.g. are not commented upon in the work 
shows either that the v.g. did not exist at the time of the writing of Nd 2, 
or that they may have existed in some form or other, but were not accepted 
as authentic by the author. The early occurrence of v.g. in verse is highly 
doubtful, but it is probable that the versification of an earlier existing prose 
legend may have taken place somewhere about the same time as the 
composition of the v.g. of the Nalaka Sutta. This introductory prose 
legend cannot be considered very old, for all the internal evidence of the 
v.g. and the epilogue shows that these pieces were at least a few centuries 
younger than the pucchas. It is probable that the legend of Bavarl which 
was introduced as an akhyana-mrmtive by the reciters of these ballads, 
underwent,certain changes and modifications as time went on, and finally 
became fixed in the present metrical rendering. The outcome is a short 
kavya in itself in true epic-style. 

The opening stanzas easily suggest their kinship with epic literature. 
A Kosalan brahmin (from Savatthi?) comes to the Southern Country 
(Dakkhinapatha of Deccan) and settles down at Mujaka (reading with 
Nd. 2 and Chalmers) on the banks of the upper Godhavarl in the country 
of the Assakas (Asmaka), probably not very far from Patitthana (Pratist- 
hana, the modern Paithan about 1 9. 5° N 75° E). 28 Then another brahmin 
visits him and demands (text, yacati— begs, Sn. 980d) 500 pieces. When 
Bavarl replies that he has no money the other curses. The pronounce- 
ment of the curse (Sn. 983), its description (Sn. 984), the repercussions on 
Bavari (Sn. 985), the appearance of the devata (Sn. 986) and the conver- 
sation that ensues (Sn. 987-993) are truly characteristic of epic poetry. 
There are numerous instances of similar situations in the Sanskrit epics 
and other literature. The pronouncement of the curse in Nalopakhyana 
and the gradual denouement of the plot in it could be compared with the 
legend of Bavari. The comparatively later jataka literature affords many 
parallels. Neumann (Reden p. 547) compares Sn. 984 with the description 
of the curse in Sakuntala. The tidings of the Buddha given by the devata 
cause immense joy in Bavarl who summons his pupils and bids them visit 
the Buddha. In reply to their question as to how they would be able 
to recognize the Buddha, Bavarl replies that he could be distinguished 
by the 32 characteristics of a mahapuri sa (super-m an ). He instructs them 
~~28. B.C. Law, in ''India as described in the Early Texts of Buddhism and Jaintem'' 
op. 157, 158, 218, tries to establish that this Bavari was Pasenadis teacher (Sn ; A. 11, 
580) and that when he built his hermitage "near the Pancavafi during Pasenadi s reign 
there came into existence a high road connecting Rajagaha and Patitthana. (ibid. p. 219). 


Pali Buddhist Review I, 3 (1976) 

not to ask their questions verbally, but merely to think of them, so that 
the Buddha would give the appropriate answer. 

The sixteen brahmins wend their way north through Patitthana to 
Mahissati (Mahlsmatl) south of the river VetravatI which divides the 
Mandate of AvantI into north and south, the north having its capital 
at Ujjeni (modern Ujjain) and the south at Mahlsmatl, 29 and known as 
Avanti Dakkhinapatha. 30 From there they proceed to Ujjeni north of 
the river and to Gonaddha. 31 They continue east to Vedisa known as 
"The Forest City" (Sn. 101 Id, Vanasavhaya, identified by Cunningham 
with modern Bhllsa in Gwalior State, 26 miles N.E. of Bhopal), and then 
north-east to Kosambi (Kaiisambl), and next north to Saketa, Setavya 
and Savatthi, the capital of the Kosalas, then eastwards to Kapilavatthu 
(Kapilavastu) of the Sakyas, and the city of Kusinara (Krslnagara) of the 
Mallas, then further north to Pava and Bhojanagara in the Malla country 
in the Himalayan foot-hills and finally south-east to Vesall of the 
Magadhas and Pasanaka cetiya near Rajagaha where they meet the 
Buddha. They are satisfied with the answers to their ' 'mental ' ' ques- 
tions and salute the Buddha. With the invitation of the Buddha to ask 
him questions to have their doubts cleared, they begin asking ques- 
tions one by one. 

The vatthu-gdtha, as a whole, depict conditions much later than the time 
of the Buddha, or even the time of the compilation of the pucchas. Inter- 
nal evidence and linguistic data show that they are decidedly later 
than the pucchas. It will be useful to analyse the internal evidence which 
consists chiefly of a study of the names of places mentioned in the story, 
the terms and technical expressions used, signs of the growth of the 
concept Buddha and the doctrinal emphasis. Firstly, the v.g. show 
intimate knowledge of the Dakkhinapatha, of far-off places like Mulaka 
(not identified) and Patitthana in the land of the Assakas north-western 
Hyderabad). The road taken by the 16 manavas was the trade-route 
running from North to South-East (Savatthi to Rajagaha). 32 The simile 
at Sn. 1014b, mahalabham va vanijo (as a merchant — longs for— great 
gain) seems to allude to the caravan-men who followed these trade-routes. 
Even if the first route did exist as early as the time of PasenadI (according 
to Sn.A. 580) it cannot be said that Buddhism had spread to these southern 
regions so early as the time of composition of the pucchas. It must have 
taken a considerable period of time before Buddhism spread to these 
regions, and places like Mahissati, Ujjeni, Gonaddha and Vedisa were 

29. Vide D. R. Bhandarkar, Carmichael Lectures 1918, p. 54. 

30. B. C. Law, op. cit. p. 104. 

31. According to B.C. Law (ibid p. 74), Mahlsmatl was later known as Gonaddha. 
But this is very doubtful and improbable. 

32. Rhys Davids, Buddhist India, p. 103. 

Pali Buddhist Review 1, 3 (1976) 


far away from the cradle of Buddhism. A knowledge of these places in 
the v.g. presupposes a time when Buddhism was known to the people in 
Dakkhinapatha even if it had not spread there. It is also of interest to 
note how the brahmins looked upon this region. Baudhayana Grhya 
Sutra V. 1 5 considers this region as unholy land. 

Suras tram Sindhu Sauviram Avanti Daksinapatham. 

Etani brdhmano gatva punah samskaram arhati. 
(It behoves a brahmin who goes to Suras|ra, etc., to perform his sacra- 
ments again), cp. Divy. 19. It is needless to say that under these conditions 
Brahmanism could not have spread to these regions very early. If that 
was so it is difficult to explain how Bavarl a brahmin, and other brahmins 
mentioned in the v.g. could find their way here to a land so far south, even 
to the furthest limits of the unholy land. If Bavarl was a historical figure 
he must have lived at a time when the brahmins had begun to consider the 
Dakkhinapatha no longer as unholy land. This fact and the knowledge 
of the trade-route to the south-west suggest that the v.g. reflect a period 
when Dakkhinapatha was well-known to Buddhist writers. The first 
time that this region is expected, with some degree of certainty, to have 
come under the influence of Buddhism, is during the reign of Asoka when 
he sent out his missionaries far and wide, Mahadeva was sent to Mahisa- 
mandela and Rakkhita to Vanavasa (cp. Vedisam Vanasavhayarn both 
presumably in Dakkhinapatha. (Mhv. 12, 3-4). 

The terms and technical expressions used in the v.g. point to a compar- 
atively late period. The use of the words visaya (in Assakassa visaye — Sn. 
977a) and mandira (in Kosalamandira— Sn. 996a, and Kusinarahca man- 
diram — Sn. 101 2d) needs investigation. The word visaya in the sense of 
region, country or kingdom may have had it?, origin in epic or Classical 
Sanskrit. It is not used in this meaning in Vedic. The nearest approach 
to it in old Pali is to be found in words like Pettivisaya or Yamavisaya (the 
realm or domain of Petas and Yama respectively). This usage in the v.g. 
appears late. The word mandira is frequently found in late Sk. in the 
sense of house or mansion, as in Pali. Here it apparently stands for a 
political or regional division. If these regions were independent kingdoms 
(or cities as in the case of the latter) they would rather be referred to as 
desa or rattha, or nagara or rajadhani. It is probable that these two 
tnandiras were two of many such mandiras within a large empire. Such 
an empire came into existence for the first time in India's history under 
Candragupta (322-298 B.C.) 33 and the next great empire was that of Asoka 
(272-232 B.C.) 33 . It may then be possible that the v.g. were written at 
least after the time of Candragupta. (Other available evidence tends 

to show that they were of still later date). 

"B. VTA. Smith, Early HistorToflndia, p. 206, assigns these dates. 


Pali Buddhist Review 1, 3 (1976) 

The next point of interest lies in Sn. 1000-1001. The sixteen mdnavas 
learn from Bavarl that the Buddha's distinguishing marks are the 32 
characteristics. Here the v.g. present a phase of development in the 
Buddha-legend, for it is not his teaching that is mentioned, but his out- 
ward marks. Sn. 1001 dogmatically states that there are two, and only 
two, courses of action open to a being who has these 32 marks on his body. 
This is further proof of a gradual crystallization that has set in. There 
are a few epithets used in the v.g., e.g. sambuddho (7 times), 34 sabbadhammd- 
naparagu (Sn. 992b), pabhahkara (Sn. 991d), vivattacchadda (Sn. 1003c) 
and andvaranadassavl, among others at Sn. 991, 992, 995, 996, 1003, etc., 
but the majority of them are found in the older parts of the prose Nikdyas 
as well. 

The phrase pubbevdsanavdsitd (Sn. 1009d) "impressed with the result- 
ant force of their former deeds" 35 too sheds some light on the date of the 
v.g. The doctrine of vasana is apparently alien to early Buddhism, though 
the same idea may be found in germinal form in phrases Mkepubbe katam 
kammam (actions done in the past). The developed idea as such is to 
be seen at Miln. 10, pubbe vasandya coditahadayo (his heart impelled by 
former impressions); Miln. 263, pubbevasitavdsand (cp. Sn. 1009d), and 
Vism. 185, katasamanadhammo, vdsitavdsano, bhdvitabhavano (he who has 
discharged the obligations of a recluse, has the resultant force of his 
former deeds impressed on him and has developed his meditations). 
Vasana is often mentioned in Nettippakarana where it occurs no less than 
12 times, 36 in a slightly different sense though fundamentally the same. 
Some suttas here are called vasandbhdgiya (pertaining to v.). All the 
works in which this term is employed reflecting on an accepted theory of 
vasana, are comparatively late. Of them the date of Vism. is to some 
extent certain; i.e. 5th century A.C. Hardy limits the date of Nett. between 
2nd century B.C. and 5th century A.C. though he is more inclined to 
favour a date in the neighbourhood of the latter limit. 37 Mrs. Rhys Davids 
in her Milinda Questions suggests a date towards the beginning of the 
Christian Era to Miln.; and in her Outlines of Buddhism p. 103, she assigns 
the date 80 B.C. These instances show that all the other references to 
vasana do not go back earlier than 2nd century B.C. This fact may, to 
some .extent, help in determining the date of the v.g. All these references 
to vasana presuppose the existence of at least, a contemporary belief in 
"former impressions". It has already been noticed that this term does 
not occur in earlier Pali works. It is probable that the concept ofpubbe- 

34. This word occurs 7 times in the v.g. and twice in the epilogue, viz. Sn. 992 a, 
994 a, 995 f, 998 d, 1003 c, 1016 a, 1031 a, 1145 c, 1147 c. There are 10 other occurrences 
in Sn. — 3 in Uraga Vagga and 7 in Mah6 Vagga. 

35. Vide P.T.S. for vasana and vasita; vasana = impression (Rhys Davids). 

36. Also vide P.T.S., s.v. 

37. Nettippakarana (P.T.S.) Introduction p. xx. 


Pali Buddhist Review 1, 3 (1976) 155 

vasana was further developed into a fuller theory by the time of the Com- 
mentaries. The frequent occurrence of this idea in Nd 2 is very significant. 
The concluding passages in the comments on each of the 16 pucchas 
contains one standard phrase in which the word vasana occurs — (vide 
Nd 2, p. xxiv), Ye tena brahmanena saddhim ekajjhd ekappayoga, ekddhip- 
paya, ekavasanavagitd. (They who were together with the brahmin, of 
similar undertakings, of similar intentions and impressed with similar 
former impressions). 

It has been noticed earlier that either the v.g. did not exist at the time of 
i he compilation of Nd 2, or if they did exist in some form or other they 
were not accepted as authentic by the writer of Nd 2. The occurrence 
of the same idea in both Ndl and the v.g. shows that neither belonged to a 
period prior to the development of a theory of vasana. The probability 
is that both the v.g. and Nd 2 were not separated from each other by a long 
interval of time, and that the subject-matter of the v.g. may have existed 
in some form before Nd 2 was compiled, and that the latter was influenced 
by it. This would explain the occurrence of the phrase ekajjhd, etc. in Nd 
2 in spite of the fact that the v.g. are left uncommented in it. In the light 
of the above observations it may be inferred that these references to vasana 
do not date back earlier than 2nd century B.C., andthat both the v.g. and 
Nd 2 which were separated by a short interval of time do not go back 
earlier than the earliest limits of the period to which Nett., Miln. and Vism. 
can be assigned; i.e. 2nd century B.C. As regards the v.g. this is further 
borne out by linguistic data. 

The v.g. contain words and linguistic forms belonging to various periods 
There are older forms lying side by side with much younger ones. These 
older forms are the same as the already existing early gatha-forms and 
belong to a stratum which is generally called ' 'the Gatha-dMect. ' ' They 
either preserve the gatha-idiom or are borrowings modelled on the langu- 
age of the gathas. There are numerous instances of younger forms, some 
betraying a strange resemblance to epic Sanskrit. It also contains highly 
developed and perhaps Sanskritic idioms and usages. Even though there 
is a preponderance of older forms, the younger forms show that these 
gathas should belong to a later period. The idioms, Assakassa visaye 
(Sn. 977 a), vast Godhavari kule (Sn. 977) are purely Sanskritic. Tasseva 
upanissdya (Sn. 978 a) is a peculiar usage, which Bdhgh. comments as 
upayogatthe c'etam samivacanam, tarn upanissdya 'ti attho," (Sn. A. 581). 
The verb vdcati (Sn. 980d) in the present tense following another in the 
past (dganchi-Sn. 979 d) is typical of Sk. epic poetry. Bhavam nanupa- 
dassati (Sn. 983b) is again the Sk. idiom though the verb is a historical 
future form. Other instances of verbs in the present tense following 
a verb in the past are at Sn. 985, ussussati and na ramati after ahu in Sn. 


Pali Buddhist Review 1, 3 {1976) 

984 d. The idiom bhotijdndti (Sn. 988a)isalsoSanskritic. Theloc. sg.,asmhn 
in asmim puthavimandale (Sn. 990 b) is very near Sk. asmin, as usually Paii 
has imasmim. Puthaulmandala as a term referring to the world belongs to 
late Sk. The words visaya and mandira have already been discussed. 
The word apacca (Sn. 991c) is seldom used in Pali although it goes back 
to Vedic apatya; but it is in frequent use in Sk. In spite of the old forms 
the sufficiently numerous younger forms are ample proof that the language 
of the v.g. taken as a whole is rather late. This is quite in accord with 
the overwhelming internal evidence which definitely shows "that the v.g. 
are of no great antiquity. 


It was stated earlier that the vatthu-gdthd were meant to introduce the 
subject but apart from the legendary introduction which has little bearing 
on the panhas (pucchas), the latter are still independent suttas. Bavarl the 
brahmin, is spoken of as the teacher of the 16 mdnavas; and in the epilogue 
Pingiya is represented as singing Buddha's praises in Bavarl 's presence 
and converting him. This, apparently, is the only connecting link between 
the legend in the v.g. and epilogue and the pucchas. Yet, a rather success- 
ful attempt has been made to incorporate in this legendary epic, the 
pucchas, and to establish a connecting thread running through the whole 
vagga. However, one loses all contact with the story of Bavarl in the 
pucchas. The Buddha is seen answering the eager questions of some 
would-be followers. Nothing else can be gathered from the pucchas- 
about these interlocutors of the Buddha, except what can be seen from 
their views and philosophical leanings. 

The position of the story of Bavarl in the Pardyana is best summed up in 
the words of E. J. Thomas, "The Pardyana is indeed old. ..There is no 
reason for thinking that this legend in its present form is of the same age 
as the Pardyana. . . It is ev'dent that even though the legend may be old, 
the same cannot be said of the details that may have been introduced 
when it was recast." 58 


Uraga Vagga 

Proceeding to the other three vaggas, the Uraga Vagga calls for attention 
next. It has already been mentioned (U.C.R. VI, 1) that the Uraga Sutta 
which has been placed at the head of the vagga has given its name to the 

Pali Buddhist Review 1, 3 (1976) 



38. Life of Buddha, p. 14. 

whole section. In many respects the opening Uraga Sutta resembles the 
Khaggavisana Sutta, but unlike the two YaM/w-ballads, Hemavata and 
Alavaka Suttas or Pardbhava and Vasala Suttas which deal with practically 
the same topic, the two are not placed together in the vagga. Both Uraga 
and Khaggavisana Suttas are didactic ballads with regular refrains running 
through them, and dealing with similar topics. Another poem which 
can be classed in the same category is Muni Sutta which resembles the 
other two in both subject-matter and style though the refrain is confined 
to only 8 out of its 15 stanzas (i.e. Sn. 212-219). All these three suttas 
are archaic in character. Available evidence suggests the independent 
existence of Khaggavisana and Muni Suttas, prior to the compilation of Sn. 
The former is commented in Nd 2 (as an independent sutta) and is quoted 
in full at Ap. I, 8-13 (Paccekabuddhdpaddnd) while the latter is mentioned 
in Asoka's Bhabru Edict as Muni gdtha, and in several other instances 
along with other sections of the Scriptures (supra). 

Dhaniya and Kasibharadvaja Suttas are similar to these three in subject- 
matter, but form a different type of ballad . They can be grouped together" 
as dialogue-ballads though the latter in reality is a narrative-ballad with 
the dialogue in mixed prose and verse. In both of them there is a great 
deal of the dramatic-element; both are didactic to a certain extent and both 
deal with farmers who eventually became lay-disciples. The former 
contains highly antithetical alternate verses uttered by Dhaniya the herds- 
man and the Bhagava respectively, while the latter in its main section 
(Sn. 76-81) contains one question by the brahmin and a long answer given 
by the Buddha in metaphors stating counterparts to some important 
Buddhist concepts, in the various implements used and actions done in 
ploughing. In both these suttas the Buddha is represented as retracing 
the very words of the interlocutors giving them a new value and a new twist 
so that the higher truths of his message are brought within the limited 
scope of a ploughman's (for herdsman's) terminology, One would 
normally expect these two suttas too to be grouped together like Nos. 6 
and 7 and Nos. 9 and 10, on account of their similarities in style and 
theme and the technique employed in them. 

The next poem Cunda Sutta differs from the first four suttas in theme 
and general tone. It presupposes a time when some monks were seen 
leading a life of evil and sin (Sn. 89). The gradual crystallisation of ideas 
regarding the ideal monk (Sn. 86) and the motive of preventing the lay 
ariyasavakas losing their faith in the virtuous monks on account of these 
evil-doers (Sn. 90) show that the poem belonged to an age of developed 
monasticism. The inclusion of this sutta here perhaps serves to connect 
the four earlier suttas of lofty ideals with the three popular suttas that 
follow The first of these enumerates the causes of man's downfall and 

: 1 

158 Pali Buddhist Review I, 3 {1976) 

deterioration (parabhava); the second details the characteristics of a 
vasala (an out-caste in the strict Buddhist sense), and the third is a 
treatise on metta (amity). The only characteristic that is commo* to 
Cunda Sutta and the two that follow it is that all three of them are dialogue- 
ballads. In the grouping together of the two suttas, Parabhava and 
Vasala may be seen signs of an attempt at some sort of arrangement of the 
suttas. Although the next sutta, Metta, is a didactic ballad it shares 
something in common with the two proceeding suttas— all three of them 
being popular in character and intended for the benefit of both monk and 
layman. Metta Sutta occurs In both Kh. (No. IX) and the Catubhanavara 
(Panttas), whereas the other two are found repeated in the Parittas only. 

The next two suttas, Hemavata and Alavaka, are of high literary merit- 
both containing the dramatic element to some extent. The fact that they 
deal with yakkhas appears to have been the reason for their being grouped 
together. The next sutta (Vijaya) contains a list of the parts of the human 
body, m poetical form. Placed last in the vagga is the old Muni Sutta, which 
probably entered the vagga last of all. 

Judging from the subject-matter, type of ballad, and the grouping of 
poems jn the vagga, it appears that this section now known as the Vrnga 
Vagga consisted of only 10 suttas at a certain stage; thus:- - 

Group I, Suttas 1-4, 

No. 5 separating Groups I and [I, 

Group II, Nos. 6 and 7, 

No. 8 separating Groups II and HI, 

Group III, Nos. 9 and 10. 

This clearly explains the position of the old Muni Sutta as the last 
member of the vagga, placedimmediately after so late a piece as the Vijaya 
Sutta. In sp lt e of its resemblance to Khaggavisana and Uraga Suttas in 
language, style and theme, it has not been grouped with them. 


Culla and Mah5 Vaggas 

The next two sections of Sn., Culla Vagga (Cvg.) zndMahd Vagga (Mvg.) 
consist of 14 and 12 suttas respectively. The total number of stanzas 
compnsing the 14 suttas of Cvg.te a little more than half that of Mvg: 
(i.e. Cvg. 183, and Mjg. 361). The majority of the suttas in Cvg. are short 
pieces whereas those of Mvg. are comparatively longer. This perhaps 
may have been the reason for naming these two sections as Culla and 

Pali Buddhist Review J, 3 (1976) 


Mahd Vaggas respectively. Yet there are exceptions as regards the length 
of the suttas in the two vaggas. The most outstanding are Brahmana- 
dhammika Sutta (No. 1 of Cvg.) consisting of 32 stanzas, Dhammika Sutta 
(No. 14 of Cvg.) consisting of 29 stanzas and Subhasita Sutta (No. 3 of 
Mvg.) containing only 5 stanzas in addition to the introductory prose. 
There are 7 suttas in Cvg. containing 10 stanzas or less, 39 and 5 containing 
a number ranging from 17 to 12. 40 The other two are the exceptionally 
long suttas just mentioned. Five suttas of Mvg. contain 32 or more 
stanzas each, 41 in addition to the prose in the majority of them; and the 
number of stanzas in six others ranges from 20 to 26. 42 The Subhasita 
Sutta which is exceptionally short for this vagga has already been mentioned. 
It is curious to note that both the long suttas in Cvg. are named Dhammika 
and that they occur as seventh and fourteenth members of the vagga. The 
fact that one of them is the last sutta of the vagga, and that they occur at 
regular intervals may suggest that they did not originally belong here. 

The suttas of Cvg. may be classified roughly into two categories: 1. 
dialogue-ballads and 2. didactic-ballads; but the classification is not 
complete by itself. On the one hand, all the suttas are didactic in some 
degree or other, but on the other, practically each sutta seems to represent 
a type by itself. Amagandha and Sammaparibbajaniya Suttas are dialogue- 
ballads entirely in verse where the interlocutor speaks but once and the 
Buddha replies with a discourse. An interesting feature is the refrain 
running through the discourse in both suttas. They deal with topics of 
general interest in all periods of the history of Buddhism. Kimsila Sutta 
also appears as such a dialogue, although the questioner's name is not 
mentioned. It is highly didactic and may equally be classed with the pure 

There are four dialogue-ballads with prose introductions. The first of 
them, Mahamahgala Sutta is highly popular in character, and the second 
Suciloma is didactic. Both these suttas introduce supernatural beings as 
interlocutors. The former contains a refrain while the latter has none. 
The next Vangisa Sutta, is an ode in the form of a dialogue-ballad. This 
is the least didactic of all the 14 suttas in Cvg. ; yet, it is by no means lacking 
in it. Here the interlocutor plays a more active part than in the other 
dialogue-ballads of this vagga. The last Dhammika Sutta is an eulogy of 
the Buddha followed by a discourse dealing with the silas and such other 
topics. There are also four didactic-ballads entirely in verse; viz. Hiri 
Dhammacariya, Nava and Utthana Suttas. Nava Sutta is named after the 
simile employed in it (Sn. 321) and the other three after their opening 

39. Nos. 3, 5, 6, 8, 9, 10 and 11. ~ " ~ 

40. Nos. 1, 2, 4, 12 and 13. 

41. Nos. 4, 6, 9, 11 {with vatthu-gatha) and 12. 

42. Nos. 1,2,5, 7, 8 and 10. 


Pali Buddhist Review 1, 3 (1976) 

words as in the case of Kimsila Sutta. The opening question in Kimsila 
Sutta can be explained as a vatthu-gdiha, although it is not specifically 
mentioned so as in Rdhula Sutta. The Rdhula Sutta differs from the above 
four in that it has two vatthu-gdthds consisting of a question and an answer, 
and ends with a concluding prose sentence. The Brdhmanadhammika 
Sutta is essentially didactic in its verse section, but it contains an introduc- 
tory prose dialogue and concludes with a confession of faith in prose. 
The opening Ratana Sutta cannot be placed in any particular category. 
It is neither a didactic poem nor a dialogue, but aparitta of later date with 
a good deal of saccakiriyd (asseveration). The Culla Vagga thus presents 
a confused mass. 

It is not quite possible to sift out the suttas that were included in the 
vagga subsequent to the formation of a vagga as such, or spot out at a 
glance the suttas on which the vagga was built later. On the whole, this. 
section as a vagga is decidedly later than the Atthaka and Par ay ana Vaggas, 
and probably later than many suttas of the Uraga Vagga. As regards 
individual poems, the occurrence of the two long suttas (Nos. 7 and 14) 
in a section of short (culla) suttas leads one to the inference that they 
originally did not belong to this vagga. One may be justified in saying 
that these two were probably either importations to the vagga or were in 
existence in some other collection prior to the formation of Culla Vagga. 
Another sutta that appears foreign to the vagga is Ratana Sutta. From 
its internal evidence and linguistic data it will be seen that it is a -compar- 
atively late poem. This, along with the fact that it occurs at the head of 
the vagga seems to suggest that it need not necessarily have belonged to 
this vagga at the outset. Neither does it follow from this that the Cvg. 
was older than these three suttas; and the question of whether the two 
longer suttas belonged to another group of suttas (vagga) before Cvg. 
came into existence will be discussed later. 


Maha Vagga 

The suttas of the Maha Vagga are a little more uniform in character. The 
Pabbajja,Padhdna and Ndlaka Suttas are narrative-ballads with occasional 
dialogue. It has already been noticed that these three suttas represent the 
earliest beginnings of a life of the Buddha in verse (U.C.R. Vol. VI, 2). 
It ; is established beyond any doubt that the Ndlaka Sutta is the same as 
the Moneya-sute of Asoka's Bhabru Edict. An analysis of Nos. 1 and 2 
of Mvg. shows that they are very old pieces. Sylvain Levi 43 identifies 

43. J. A. 1915. Regarding Bimbisarapratyudgama he says, "Le P'in-po-chd-lo- 
po-lo-cha-k 'ia-mo-nan, < Bimbisara vient au-devant* est sans doute le Pabbajjdsutta du 
Sutta Nipdta. ' ' cp. Mvastu. 11. 198, Vinaya of the Mulasarvastivadins, Sanghabheda- 
vastu chap. 4: Vinaya of the Dharmaguptakas, Upasampadavastu chap. 31. 

Pali Buddhist Review 1, 3 (1976) 


Pabbajja Sutta as being mentioned in the Vinaya of the Sarvastivadins 
(chap XXIV) in a passage which refers to other texts as well, which he 
considers are of great importance for the history of the Canon. Of the 
other nine suttas, eight (except Salla, No. 8) are < '^"ballads. Seven 
of them are dialogues. It has been pointed out that No 3, Subhastta Sutta 
is too short for a section of "long" suttas (maha) Nos. 4-7, 9 and 10 
are uniform in every way. No. 8 is a didactic-ballad deriving Us name 
from the oft-used metaphorical term salla occurring m stanza 19 (Si,. 592). 
The Dvayatanupassana Sutta stands as a class by itself in the whole of Sn. 
It conveys the general impression of a late sutta. Its posit™ , as last* 
the vagga, as in the case of Ratana Sutta which is at the head of Cvg seems 
to strengthen the supposition that it was an additional accretion, though 
hs lateness is not necessarily proved thereby. Evidence for its lateness 
is to be sought in the sutta itself. 

The majority of the pieces in Mvg. can be called "mixed-ballads'' with 
dialogue- viz. Nos. 3-7, 9 and 10. Six of these, including No. 3, Subhasita 
Sutta k* best described as ' 'sutta- ballads," i.e. they are disccurses in the 
form 'of mixed-ballads-and the latter is more in the nature of an exposi- 
tion (veyyakarana), rather than a ballad proper. There are also four such 
^-ballads" in Cvg. viz. Nos. 5, 7, 12 and 14, which occur in a regular 
pattern in the vagga. (Suttas 5 and 12 resemble each other in outward 
form- both are short mixed-ballads with dialogue, though fundamentally 
the latter * an ode followed by a discourse, while the former a didactic 
discourse in answer to a question. The pair Nos. 7 and 14 has been 
discussed at length.) The suttas 6 and 13 too resemble each other m 
S^SpoU bSth being entirely in verse. The only difference between 
the, two ifthat the former is a straight-forward didactic poem while the 
latter is a didactic discourse in answer to a question; but the two are 
iar in outward form. The symmetry seen in these three pairs of 
suttas cannot be a mere accident. It seems likely that m building up the 
Culla Vagga these suttas have been so placed as to work out a definite 


This leads to the question whether these suttas belonged to some otner 
group or vagga before Cvg. came into existence. If there was any such 
Souo some of the suttas now found in Mvg. should also have been includ- 
fd m it for, the existence of a section called Maha Vagga without a corres- 
nondine Culla Vagga is very doubtful. 44 The resemblance of suttas 4-7, 
ttorflS the Z "mixed-ballads" of Cvg. in form and style 
Lgests that they too may have been included in such a group. There is 
notnS ; to prevent No^ofJ^emgJn^ 

too numerous and need no mention here. Vide D.P.P.N., Maiaias 


Pali Buddhist Review 1, 3 (1976) 

Pali Buddhist Review 1, 3 (1976) 


be argued that as No. 9 in Mvg. is rather expository in character, No. 8 
which is an expository didactic poem should have been placed in front 
of it as in the present vagga. But the greater probability is for the same 
type of "mixed" dialogue-ballads to be grouped together, like the pucchas 
of the Parayana. This would mean that the hypothetical vagga consisted 
of Cvg. 5, 7, 12 and 14 and Mvg. 1-7, 9 and 10. 

This reflects no light whatsoever on the question of the age of the suttas 
found in. these two vaggas. Beyond any reasonable doubt Moneyya Sutta 
(Nalaka discourse) could be placed among the oldest suttas in Sn. The 
age of the suttas does not necessarily determine whether they belonged to 
a particular group (or groups ) or not, for, they can exist independently 
and be introduced into other collections at subsequent dates; e.g. the old 
Muni Sutta, a comparative new-comer to Uraga Vagga. This further 
justifies the exclusion of old suttas like Pabbajjd, Padhana and Nalaka 
from the reconstructed group of ballads. Moreover the position of 
these suttas in Mvg. indicates that they were probably additions made 
when two vaggas grew in place of a vagga of mixed-ballads. (This need 
not necessarily have belonged to Sn., and its independent existence like 
the Parayana or A tthaka Vagga is not improbable). Pabbajjd and Padhana 
Suttas were placed at the head of Mvg. (and not Cvg.) probably on account 
of their length. The only plausible explanation of the position of the 
short Subhasita Sutta as the third member of the vagga is that it could 
have occurred in some collection or other together with the preceding 
suttas; but this is highly improbable. As it differs considerably from the 
' \sw/to-ballads ' ' it cannot be surmised that it may have occurred immedi- 
ately before Sundarikabharadvaja Sutta in an earlier group. As regards 
the Salla Sutta, its length and the expository nature of the following 
(Vdsettha) sutta may have been responsible for its inclusion in the present 
Maha Vagga, and probably it did not exist together with the others in an 
earlier group. The Nalaka Sutta seems to have been introduced immedi- 
ately after the regular ' 'mixed-ballads. ' ' The chief reason for its inclusion 
here and not in Cvg. is its length. One would normally expect this sutta 
to be placed beside the other two suttas which are directly connected with 
the life of the Buddha. The fact that this is separated from them also 
suggests that these three suttas did not originally belong here, but were 
introduced after the two groups Mvg. and Cvg. were formed. 

It may also be possible that the three suttas, Pabbajjd, Padhana and 
Nalaka were earlier found together in one group at a certain stage, and 
that eight suttas were added after the Padhana Sutta to make up the Maha 
Vagga. The fact that these three suttas belong to an early stratum does 
not necessarily imply that they may have been the only suttas of their 
class. Moreover, Nalaka Sutta does not form a continuous narrative 

with the other two suttas. A comparison with the later BSk. sources, 
such as Lai. which aims at dealing with a continuous life of the Buddha, 
or Mvastu, which contains accounts of incidents connected with his life, 
shows that these three suttas in Sn. deal with only three of the numerous 
incidents reported in later sources. It is quite probable that some suttas 
parallel to those found in Lai. and Mvastu, were lost and that Sn. contains 
only a partial picture. The fact that only these three are preserved shows 
that they are but fragments of an earlier stratum brought to light at a 
subsequent date and included in the group now known as Maha Vagga. 
It has' already been pointed out that their relative position in the vagga 
shows that they are additions made to the vagga rather than parts of its 
framework. From these it is evident that Mvg. was not built upon these 
suttas but it grew incorporating them. 

It is not possible to determine whether any one of these two vaggas was 
earlier than the other (as a vagga). Neither of them is a perfect ' 'finished" 
chapter. Though the majority of the suttas conforms to the designations 
Culla and Maha, in length, many exceptions have already been noted. 
The themes in the "minor" suttas (i.e. those in Cvg.) are equally lofty as 
those of the suttas in Mvg. Therefore the possibility of the two sections 
being named according to the nature of the themes can be set aside. 
There is no perfect uniformity in the type of suttas in both vaggas though 
as many as six suttas of Mvg. can be classified as "mixed" dialogue- 
ballads. The same type of sutta is to be seen in Cvg. too; viz. Nos. 4, 5, 
7, 12 and 14, though the didactic element seems to predominate in them. 
The commonest type of sutta in Cvg. is the pure didactic-ballad entirely 
in verse. 45 but Mvg. No. 8 (Salla Sutta) too can be said to belong to the 
same type. The similarity of these two vaggas even on this point suggests 
that they cannot be separated from each other in point of time. Both 
vaggas date back to the same period, and the occurrence of the older 
suttas in Mvg. proves nothing beyond the fact that they were incorporated 
into the vagga during the time of its compilation, which perhaps was 
synchronous with the collation of Sutta Nipata as an anthology. 


45. There are seven such suttas; viz. Nos. .?, 6, 8. 9, 10, II and 13. 


N. A. Jayawickrama 
Additional Abbreviations 

Aor. — aorist 

Brh. Kv.—Brhadaranyaka Upanisad 

Dh.—Dhammapada (PTS, 1914) 

E./W.Pkr.— Eastern/Western Prakrit 

J—Jataka (PTS, 7 vols., 1877-97; reprinted 1962-64) 

JA—Jataka Commentary (ibid.) 

Mbh. — Mahabharata 

P.— Pali 

PBR— Pali Buddhist Review 

•pvz—ParamatthadipanKPetavatthu Commentary, PTS, 1894) 

Yt%A—Saddhammappakasini(Patisambhidamagga Commentary, PTS, 3 

vols., 1933-47) 
S.K.—Saratthappakasini (Samyutta Nikaya Commentary, PTS, 3 vols., 




It is now possible to proceed to the analysis of a few individual suttas 
of the Sutta Nipata with the aid of the criteria detailed earlier. The 
following analysis is restricted to a proportionately small number of 
suttas and further inferences regarding those that are left out may be drawn 
on similar lines. Every opportunity will be taken to discuss problems of 
general application to the whole work under the discussion of these suttas 
so that most of the problems connected with the majority of the suttas, 
will be eventually touched upon. An attempt has been made to make the 
selection as representative as possible. A few suttas from each vagga 
and from each type in the classification on pp. 88-90 in PBR 1, 2, are 
taken up for analysis. Wherever possible the suttas will be discussed in 
the order in which they occur in Sto.,and at the same time those that bear 
some similarity to one another will be arranged in some order so as to 
bring out the properties they share in common. 

Pali Buddhist Review 2, 1 (1977) 


The title Uraga Sutta is undoubtedly derived from the simile, urago jin- 
nam iva tacam puranam (as a serpent — discards — its old and worn-out 
slough) in the refrain that runs through the whole length of the poem. 
Its ability to cast off its slough, an important characteristic of the uraga, 
has been introduced here to describe the action of the bhikkhu who re- 
nounces both ' 'Here and the Beyond. ' ' There seems to be some myster- 
ious significance attached to this creature which is described as ura-ga 
(lit. belly-crawler). Some uragas are considered to belong to a class of 
semi-divine beings: they are kama-rupi (SnA. 13, capable of changing 
their form at will). The semi-divine characteristics are usually attributed 
to nagas rather than to uragas. There are numerous instances in the Pali 
Canon of nagas changing their form or appearing in disguise. The 
Commentary (SnA.) refers to Sankhapalanagarajd in Sankhapalajataka 
JA. V. 161-177). At Vinayal, 86 a naga is said to have received ordination 
disguised as a young man. An equally mystic significance is associated 
with the uraga's casting offof the slough. The Commentary (SnA. 13-14) 
describes in detail the four ways in which it does so. PvA. 61-62comment- 
ing on Pv. I. 12, 1, urago va tacam jinnam hitva gacchati sam tanum (he 
goes abandoning his body—corporeal form—as a serpent discards its worn- 
out slough) says that a serpent casts it off whenever it wishes to do so, 
as easily as removing a garment, with no attachment to it whatsoever. 
Here the simile of the serpent's slough is employed to describe the body 
at death. The mysterious significance of the uraga is more pronounced 
in a passage occurring at SI. 69. It describes four young creatures 
(dahara) which should not be despised nor abused viz. a khattiya, an uraga, 
aggi and a bhikkhu. A khattiya when he becomes king can inflict heavy 
punishment on man, woman or child that despises him; an uraga can sting 
them; therefore he who holds his life dear should not despise it. Fire 
with necessary fuel (upadana) can blaze forth into a huge flame and burn 
them who despise it. The virtuous bhikkhu can burn with his flame-like 
majesty. The uraga is also described as, uccavacehi vannchi urago carati 
tejasi (v. 1. tejasa) in the Samyutta: (In diverse appearances 1 the uraga 
roams in its own splendour). It is described here as a mysterious and 
wonderful creature demanding respect and adoration. The Commentator 
is silent about the pada, urago carati tejasi, and does not confine the 
quality of tejas to uraga alone. Fire too possesses the same quality. 
The tejas (splendour or better, power) of the uraga is perhaps due to one 
or more of the following reasons: — 

1. Cp. Corny. S.A. I, 132 nSnavidhehi san)hanehi, etc. 


Pali Buddhist Review 2, 1 (1977) 

1. Its extraordinary force or strength. 

2. Its ability to change at will (kdmarupa). 

3. The ease with which it casts off its slough. 

4. The fact that it possesses two tongues (dujivho cp. 7.V. 425 cp. II. 

458, and III. 458). 

5. Perhaps its ability to live even in fire, like the mythical salamander. 2 

Uraga is almost synonymous with ndga, a creature equally gifted with 
miraculous powers and great strength. Some of them are semi-divine. 
Ndga is often used as an epithet for arahants and sometimes of the Buddha. 
Popular etymology explains ndga as "dgum na karotV (cp. $n. 522a); 
and the origin of the epithet is perhaps based on the great power of the 
ndga. The phrase hatthi-ndga suggests an equally mystic significance. 
Yet, it is noteworthy that the word ndga is hardly or never used in the Pali 
Canon in the same simile of its shedding the slough. Though sappa, ahi, 
dsivisa and bhujangama are synonyms for uraga, they fall short of the 
connotation of the latter term. There is no real magical power attributed 
to them, unlike the uraga or the ndga. The last of the four synonyms 
bhujangama, though not in frequent use in canonical Pali seems to be 
nearer uraga than ahi. Sappa is treated as a mere poisonous snake in 
similes. It should be avoided; e.g. Sn. lb, 768b, Th 1. 457, and /.V.18. 
Asivisa is employed in similes to describe kdma and similar evil tendencies; 
e.g. Th 2. 451, /. III. 525, cp. 267 and S. IV. 172-174. 3 It is called uggatejas 
at S. IV. 172 ff. and is a synonym for the mahddhdtus. 

It has already been remarked that the emphasis on uraga in the refrain 
has been the basis of the title Uraga Sutta. There are three other suttas in 
Sn. named after a simile or metaphor occurring in them; viz. I. 3 (Khagga- 
visdna), II. 8 (Ndvd) and III. 8 (Salla). Over half the number of suttas in 
Sn. are named after the interlocutors mentioned in them. There are 36 
such suttas; viz. Sn. I. 2, 4, 5, 9, 10; II. 5, 11, 12, 14; III. 4, 5, 6, 7, 9, 10, 11; 
IV. 7, 8(?), 9, 16; and V. 2-17. To this may be added I. 7 which is called 

2. Here fire and the uraga both possess tejas. A. K. Coomaraswamy in his Hinduism 
and Buddhism makes out that the Dragon is the sacrificer and the sacrifice, i.e., 
connected with the smoke coming from the sacrifical altar. Referring to S.B. I. 6.3.14 
ff, he attempts to identify the Dragon with the Projenitor. Some such mystical signific- 
ance may be among the reasons for ascribing tejas to the uraga. 

At Vin. IV. 108 a naga (Vin. I. 24, nagaraja) is described as iddhima: but in each case 
he was overcome by the (greater) tejas of his opponent (Sagata and the Buddha) cp. J.I. 
360. It is also told at AA. I. 324 ff. how Sagata tamed the fierce nagaraja. (Note by 
Miss I. B. Horner). 

3. Vide Mrs. Rhys Davids: Similes in the Nikayas, J.P.T.S. 1906, pp. 52 II., 1908 
pp. 180 ff. 

Pali Buddhist Review 2, I (1977) 17 

Aggikabhdradvaja Sutta in the Commentary. Of the above list Pasura 
Sutta is rather doubtful, for Pasura may not be a proper name as Neumann 
(Reden p. 528) suggests. He favours the commentarial gloss pali-suro 
and says that it is pa-sura (pra-sura) like pdcariya at M.I. 509. There are 
15 suttas named after the topics or themes discussed in them; viz. Sn. I 6 
7, 8, 12; II. 1, 2, 4, 7, 13; III. 1, 2, 12; and IV. 1, 6, 11. To this may be' 
added the alternative names given in the Corny, for 1. 1 1 (Kdyavicchandanika) 

II. 8 (Dhamma), II. 12 (Nigrodhakappa), II. 13 (Muni or Mahdsamaya), 

III. 4 (Puraldsa) and III. 1 1 (Moneyya). In addition to the six suttas men- 
tioned in the note on p. 79 in PBR 1, 2, as being named after their opening 
words, Sn. IV. 10 (Purdbheda) is named after the opening word of the 
second stanza (i.e. Buddha's reply). The four Atthakas have already 
been mentioned (PBR 1, 3, p. 143) to contain in their opening lines the 
words after which they are named. This makes a total of 1 1 suttas that 
are named after an opening line. The titles of 9 of these suttas (i.e. except 
Purdbheda and Attadanda) have direct bearing on the topics discussed in 
them. The other four suttas in Sn. viz. I. 11, IV. 12, 13 and 14 are given 
descriptive titles. It is significant that all the four suttas named after a 
simile occurring in them are pure didactic ballads and all the suttas named 
after persons are dialogue ballads. Those that are named after topics 
discussed in them belong to various types. There are dialogue ballads 
like Vasala and Brdhmanadhammika Suttas, didactic poems such as Muni 
and Kama SWta^narratives like Pabbajjd and Padhdna Suttas and doctrinal 
dissertations such as Dvayatdnupassand Sutta belonging to this group. 
All the suttas, named after their opening words are didactic poems. 

Coming back to the Uraga Sutta, the effectiveness of the simile of the 
serpent 's skin may have been one of the reasons for placing this sutta at 
the head of the vagga, which in turn derives its name from the former. 
This is the only vagga in Sn. which is named after a sutta. However, the 
practice of naming vaggas after suttas is not rare in other parts of the 
Canon. There are two Yodhdjiva Suttas occurring in the Pancaka Nipata 
of the Anguttara, viz. HI 89 ft. and 93 ff., and the vagga in which they 
occur is called Yodhdjiva (III. 84-110). Similarly, the second vagga in 
the Majjhima, Sihandda, I. 63-122) is named after the two opening suttas 
Culasihandda and Mahasihandda (Nos. 11 and 12) and the eleventh vagga 
{Devadaha, II. 214-226, III. 1-24) derives its name from the opening 
Devadaha Sutta (No. 101). In such instances as these it need not be the 
opening sutta that is always responsible for the name of the vagga. In 
the Uddna, the third vagga, (Nanda, Ud. 21-33) derives its name from the 
second uddna in it, its fifth vagga Sonatthera, Ud. 47-61) from its sixth 
member, its sixth vagga (Jaccandha, Ud. 62-Th) from the simile in the fourth 
piece in it, and the last vagga (Pdtaligdma, Ud. 80-93) from the sixth 



Pali Buddhist Review 2, 1 {1977) 

udana in it. The first two vaggas of .Pv. are named after their concluding 
members. Uragapetavatthu and UbbaripetavatthuTespectb/ely. Of them 
the Uraga Vagga is of special interest here. The first verse of the vatthu 
at Pv. 11 (I. 12. 1) contains the simile of the serpent's worn-out slough, 
and this is the only reason for naming the vatthu and the vagga, Uraga. 
The Corny, associates this vatthu with an uraga (serpent) which was 
responsible for the death of the individual referred to in the story. The 
illustrative story in the Corny, is the same as that at J. III. 162-168, which 
also contains the text at Pv. I. 12 in full. 

The simile of the snake casting off its slough seems to be rather popular 
in Pah verse. The line at Pv. I. 12 1 is also found at Ap. 394, 13. In 
Mora Jdtaka (J. IV. 341) the hunter renounces his career as a hunter even 
as a serpent discards its old worn-out skin (tucam vajinnamurago purdnam). 
Pv. IX. 28 contains the same line. This simile is also employed to 
describe how Fortune keeps the fool at bay at J. V. 100 and VI. 361: 

Sirijahati dummedham jinnam va urago tucam 


The 17 stanzas of the text describe the bhikkhu who overcomes anger, 
lust, craving, arrogance, hatred, doubts and perplexities and other impedi- 
ments, has found no essence (sard) in all forms of being, sees everything 
as void being free from covetousness, passion, malevolence and delusion, 
has eradicated all evil tendencies with no leanings whatsoever towards 
them, is free from all such qualities which form the basis for earthly 
existence, and has destroyed all obstacles. He verily is "the bhikkhu 
who shuns both Here and the Beyond as a serpent its old and worn-out 

The tone of the sutta is generally archaic and the language preserves an 
early stratum of Pali. The words and forms of interest are: — Oraparam 
(Sn. lc-17c), a simple dvandva cpd. meaning "here below -cp. Sk. avara- 
and the beyond," cp. parovaram (Sn. 353, etc.). The ora and the para 
are the limitations (sima) to a true bhikkhu. If he wishes to go beyond 
them (slmdtigo, cp. Sn. 795a) he should rid himself of all obstacles and 
leanings which act as causes (lit. causal antecedents) for his downfall (cp. 
Sn. 15b). The concept ora has already been noted (PBR 1, 3, pp. 147-8) 
as being the opposite of para; but para in this context is quite different 
from that of the Pdrdyana and other places in Sn. Here it merely denotes 
birth in other existences whereas elsewhere (loc. cit.) it is almost a synonym 
for nibbdna. Udacchida (Sn. 2a, 3a) cp. Sk. ud-a-chid-at; augmented 
radical Aor. 3 sg. cp. Vedic. The change -a+ch->-acch- is due to 

Pali Buddhist Review 2, 1 (1977) 


metrical reasons and for preservation of quantity. This is a pure gdtha- 
form not met with in canonical prose. There are four other such aug- 
mented radical Aor. forms in this sutta: viz. udabbadhi (Sn. 4a), ajjhagama 
(Sn. 5a), accagamd (Sn. 8b) and accasari (Sn. 8a- 13a). This type of Aor. 
is very frequent in Sn. and other old (gdthd-?a\i? Udabbadhi is usually 
explained as being formed from ud+\/vadh cp. udavadhit. It is probable 
that this verb is associated with ud+y/vrh, brnhati, to tear, cp. A.M. 
abarhit and Brh. Ar. avrksat, P. udabbahe (Sn. 583b, opt. 3, sg.) and 
abbahi (Aor. 3 sg.) in the phrase abbahi vata me sallam used frequently in 
Th 1 and Th 2. The probable development of udabbadhi from ud+^vrh 
is as follows: — Vedic udabarhit ud-a+v/b (v. in vrh)>¥. udabb-: Vedic 
-h-> P. -dh- cp. Vc. iha>P. idha. It may be possible that this form is the 
result of a contamination of the two roots vadh and vrh but either 
of the two can give this form directly and makes the explanation of a 
contamination superfluous. Ajjhagama (adhi-a-gam-at) and Accagama 
(ati-a-gam-at) cp. abbhidd (J. I. 247), dsada (Th 1.774), acchidd (Sn. 357c) 
and udacchida (supra). Accaszrl (occurring in both ndccasdri and paccasdri) 
ati-a-sar-it, from \/sr, sarati. Neumann (Reden, p. 406) suggests atyasmdri 
and pratyasmdri, but the explanation in the Corny., na atidhdvi and na ohiyi 
is preferable. SamOhatase (Sn. 14b) cp. paccayase (Sn. 15b) ; double Vc. 
nom. pi. from-dsas>-dso>-dse; -o>-e is a dialectical variation influenced 
by E. Pkr. (Magadhi). This double nom. with Magadhi -e is rather 
frequent in early Pali poetry. In Sn. alone it is seen to occur 20 times, 10 
of which are in the Atthaka Vagga. Neumann (Reden, p. 407) says that 
such forms as samu-hatdse are not "Magadhisms'" but periphrastic 
perfects; samuhata+dse. It would be rather straining to construe a 
perfect in such contexts as these, and his suggestion, however useful, is 
not tenable. HetukappS — "which act as a cause " (Sn. 1 6b). The cpd. 
appears to be dialectical and nearer the older language, cp. khaggavisdna- 
kappo" "resembling a rhinoceros" (Sn. 35d-75d). 


The Uraga Sutta is written in a metre described as Aupachandasaka 
by Helmer Smith (SnA. 463). The regular Aupacchandasika metre differs 
from the Vaitdliya which consists of two half verses with 30 morae each, 
in that it has an extra long syllable added to each line of 14 and 16 morae 
respectively in the Vaitdliya. The metre of these stanzas is rather irregular. 
The number of morae in the first half-verse varies from 32 to 36, but the 
average seems to be 33, as in the case of the common second half of all 
these stanzas. The extra syllables in the longer lines may be explained 

4. Vide Geiger, Pali Literatur unci Sprache, 1 59 ff. 


Pali Buddhist Review 2, 1 (1977) 

as anacrusis. Helmer Smith (ibid.) further notes padas Sn. 6b, 7a, 8a- 13a 
and 14b as irregular. Though this metre is similar to Aupacchandasika 
which became fixed in the Classical period of Sanskrit literature this 
particular rhythm cannot be considered so late as that, for it may have 
been in use long before a metre as such came to be fixed. 

Another noticeable feature here, as well as in all Pali poetry is the 
apparent disregard of metrical rules. This probably may be the result 
of the composers being guided more by the ear (rhythm) than by such 
artificial means as fixed metres. Moreover, in all popular poetry metrical 
rules are not strictly observed. However, the beat and rhythm of these 
lines resemble those of dance metres which are usually free and easy 
metres not subject to artificial regulations. 

The style of this sutta has already been commented upon. It is a ballad 
in every respect, though it is used for a didactic purpose. The purpose 
of the refrain in lines cd in each stanza is to lay emphasis on the central 
theme. There is a refrain in the initial line and the greater part of the 
second line of stanzas 8-13, 

Yo naccasarl na paccas&ri 
sabbam vitatham idam ti vita — 

There is perfect antithesis in the two halves of all these stanzas. Invari- 
ably the stanza begins with yo and the second half with so- bhikkhu. In 
spite of this and the lucidity of diction there is no poetic extravagance 
which characterises later compositions. Popular similes are freely used 
to describe the bhikkhu who leads a life in accordance with Buddhist 
ethics. Neumann (Reden, p. 408) points out a few parallels in Mbh. and 
other early literature, viz. jirnam ivacam sarpa ivavamucya (Mbh. V. 39, 
2; cp. XII. 250, 1 1) and yathd pddodaras tvacd vinirmucyate (Prasnopanisad, 
5, 5; also Wife Brh. Ar. IV. 4.10). Other similes are at Sn. lb, 2b, 4b, 
(compared by Neumann with Rgveda I, 32, 8) and 5b. 

There is nothing extraordinary in thought and ideology in the sutta. 
The emphasis is on the conduct of the bhikkhu. It is noteworthy that there 
are 80 references to bhikkhu in the gathas of Sn. (in addition to over 1 5 in the 
prose), 77 to muni (24 of which refer to the Buddha) and over 40 to samana, 
at least 17 of which are used without any specific reference to a Buddhist 
satnana. The Sangha is mentioned 8 times in the Ratana Sutta and 4 
times elsewhere in both prose and verse. All the references to bhikkhu, 
muni and samana amply justify Fausboll's statement "we see here a 
picture not of life in monasteries but the life of hermits in its first stage." 5 
The Uraga Sutta like Tuvataka and Sammdparibbdjaniya Suttas is a splendid 

5. His translation of the Sutta Nipata, p. xii (SBE, vol x). 

Pali Buddhist Review 2, 1 (1977) 21 

example of a poem that describes the true bhikkhu just as Muni Sutta 
describes the muni. 

The few technical expressions used indicate a phase of development in 
the doctrine. The terms ora and oraparam have already been discussed. 
The terms kodha, raga, tanha, mdna bhavesu sdra (cp. bhavatanha), kopa and 
vitakka(Sn. 1-7) have not undergone the later systematization and arrange- 
ment in groups. Usually the three raga, dosa and moha occur in one 
group in the more systematised texts and are called the three akusala- 
muldni (fundamental blemishes of character). At some places kodha and 
upanaha are added to these three, while at others kilesa and kodha and 
still others mana together with or without ditthi. Similarly the occurrence 
of papanca, vitathd, lobha, raga, dosa and moha in Sn. 8-13 seems to 
presuppose a time prior to the scholastic classification of the three akusala- 
mulani as raga/ lobha, dosa and moha. Besides Sn. 14 seems to associate 
mula akusala with anusaya. 

On the other hand the technical significance of daratha and vanatha 
(Sn. 15 and 16 respectively), the mention of panca nivarana (Sn. 17), and 
the emphasis on imam papancam (Sn. 8) seem to suggest a development in 
terminology. These are the only references to them, in the form as they 
are, in this work, though Sn. 514d mentions the nivarandni and Sn. 66a 
the pancavaranani. This seemingly developed terminology may probably 
point out that this sutta presupposes a time when some form of system- 
atization and arrangement has just set in. Another interesting word is 
itibhavabhavatam (Sn. 6) which is translated by Fausboll as 'reiterated 
existence' and by Neumann as 'being and non-being'. The explanation 
in the Corny. (Sn.A 20), sampallivipatti-vuddhihani-sassatuccheda-punna- 
papavasena iti anekappakara bhavabhavata does not make it clear at all. 
It seems to be somewhat different from bhavabhava which occurs in 8 
other places in Sn. 6 in the sense of 'reiterated existence' or re- birth. Its 
meaning in this context is apparently nearer the idea of the fluctuating 
changes of fortune in the course of re-birth. The term is not strictly 

An examination of Uraga Sutta in the various aspects of language, 
metre, style, doctrine and ideology shows that it is a comparatively old sutta. 
The lack of linguistic forms that may be classed as late and the presence 
of old Vedic and dialectical forms suggest that the sutta preserves an old 
stratum of Pali. The syntax of the stanzas is also simple. The flexibility 
of metre also suggests an early date for the sutta. The lucid and simple 
style which is by no means heavy or laboured is characteristic of old poetry. 
The doctrinal emphasis too speaks of an early date for the sutta; and the 

6. Sn. 496b, 776d, 786d, 810b, 877d, 901d. 1,060b and l,068d. 


Pali Buddhist Review 2, 1 (1977) 

few technical expressions reflect the ' 'germs of a philosophical system 
which came to be more logically and consistently systematised" 7 later on. 
Nothing could be gleaned regarding the social conditions of the time from 
this sutta; all other available evidence points to an early date. 


This sutta like the Uraga Sutta derives its name from the simile used 
in the refrain. 

eko care khaggavisdnakappo 

(let him wander alone like a rhinoceros). The lonely habits of the rhino- 
ceros are symbolic of the solitary wanderings of the ascetic — muni. Rhino- 
ceroses like elephants expelled from the herd are known to lead a solitary 
life. Yet, there seems to be some disagreement about the title which is 
often rendered as ' 'The Horn of the Rhinoceros ' ' following the explanation 
in the Corny., ettha khaggavisdnam nama khaggamiga-singam (khaggavi- 
sdna in this context means the horn of the rhinoceros — SnA. 65). This 
explanation may be accepted on the mere coincidence that both species 
of the rhinocerbs seen in India, viz. the "Indian" and the "Javanese" 
possess only one horn, 8 and that the animal itself is called khagga in Pali 
and khadga in Classical Sanskrit. The explanation of khaggavisanakappo 
at Ndl. 129, yatha khaggassa nama visdnam ekam hoti adutiyam. .(just 
as a rhinoceros possesses only one horn and not a second...) also justifies 
the explanation in the Corny. In spite of all this the simile would be 
considered more apt if the life of the lone-sojourner was compared with 
the lonely habits of the rhinoceros than with its single horn. 

In other places in the Pali Canon the idea of wandering alone is com- 
pared with the movements of animals of solitary habits rather than with 
parts of their anatomy. The simile employed at J. II. 220 is with reference 
to an elephant that wanders alone — gajam iva ekacarinam. The simile, 
eko care mdtahg-arahhe va nago (let one wander alone as an elephant in 
the forest frequented by mafrwiga-elephants) at M. III. 154, Dh. 329, 330 
and /. III. 488 cp. V. 190 too makes it clear. The similes, migo arannamhi 
yatha abaddho yenicchakam gacchati gocardya (as an untethered deer in 
the forest-glade roams at will for pasture) at Sn. 39ab, and nago va yuthdni 
vivajjayitvd (as an elephant that forsakes the herds) at Sn. 53a can be 
compared with that in the refrain. It will be clear from these examples 

7. B. C. Law, History of Pali Literature, Vol. I, p. 239. 

8. Sub voce Encyclopaedia Brittanica. 

Pali Buddhist Review 2, 1 (1977) 23 

that the point of contact of the comparison is an action (i.e. the wandering 
—cariya) and not an object. Moreover, even in the similes employed in 
the poem where inanimate objects are compared it is rather some action 
that stands for comparison than those objects; e.g. vamsdkaliro va asajja- 
mano (like a bamboo-shoot not clinging to anything) at Sn. 38c, samslna- 
patto yatha kovildro (like a kovilara tree with its scattered- leaves) at Sn. 
44b, aggi va daddham anivattamano (like fire not returning where it had 
burnt) at Sn. 62c and sanchinnapatto yatha parichatto 9 (as a pdrichatta 
tree with its leaves cut off) at Sn. 64b. 

From these examples it is rather convincing that the point of contact 
in the simile of the khaggavisdna is not khaggassa visdna (rh. 's horn) nor 
the cariya (movement) of the visdna (horn) of the khagga; but the cariya 
of the khaggavisdna, the sword-horn (the rhinoceros) itself. It is quite 
probable that the rhinoceros was known in earlier Pali as khaggavisaria— 
that which possesses a sword-like horn 10 and that the term khagga came 
into usage later on. This is further testified by the few comparatively late 
passages in which the animal is called khagga viz. Ndl. 129, SnA. 65, JV. 
406, 416, VI. 277 and 538. It would therefore, be more correct to interpret 
the word khaggavisdna a.& "rhinoceros" and not "rhinoceros' horn". 


The sutta on the whole deals with a life of solitude. It advocates the 
cessation from attachment to family life, friends and companions and 
society in general. The refrain eko care is employed to exhort one to 
adopt a life of solitude. The idea so colourfully painted in the simile is 
stressed over and over again in other similes. 11 All the stanzas are con- 
nected with the central theme, yet in certain places the connecting thread 
appears rather thin. A few apparent repetitions and the interruption of 
the logical trend seem to suggest that the present sutta is an enlarged 
version of an earlier nucleus. It is of interest here to no{e that the Khad- 
gavisdna Gathd at Mvastu. I. 357, consist of only 12 stanzas. A com- 
parison of the two versions shows that both deal with the same topic and 
that the BSk. sutra, though short, discusses the question of solitude as 
fully as the Pali version with all its digressions and apparent contradictions. 
While the central idea of the Pali sutta is the giving up of friends and 
.companions, sons and household life and all forms of samsagga and 
santhava (ties and attachments), there are occasional references to an 
"ideal companion" 12 an idea which appears to be an importation to the 

9. Cp. Mvastu. I. 258, saniiirna-patro (with scattered-leaves). 

10. Cp. English, horn-bill, sword-fish, etc. 

. 11. At Sn. 38, 39, 44, 46, 53, 62, 64, 71 and 72. 

12. At Sn. 45, 47 and 58. 


Pali Buddhist Review 2, 1 (1977) 

original sutta. There is also other extraneous matter such as the mention 
made of certain recluses (paribbajakasl) who are virtually householders 
(Sn. 43ab), the reference to the theory of metta at Sn. 42a and the four 
items of the brahma-viharas at Sn. 73, a digression on kama and other 
upaddavas (hindrances) at Sn. 50-56 — though the stanzas conform to the 
central theme, the repetition of the idea at Sn. 46 in different words at Sn. 
57 thus re-introducing the topic of mittam ularam (a noble companion), 
and the introduction of a complete list of Buddhist terms at Sn. 60. 
Besides these, there are numerous repetitions of ideas and wholesale 
lines and phrases. 

The 12 stanzas in Mvastu. roughly correspond to 7 stanzas in Sn. in the 
following manner:— St. l/JSn. 68, st. IffSn. 73, st. 3abd//5n. 35abd, st. 
3c// ? Sn. st. 4// Sn. 64, st. 5abd// Sn. 62abd, st. 5c// Sn. 64c, stt. 6abd-10abd 
// Sn. 36abd, (st. 6c//? Sn. 36c. st. 7c//? Sn. 37c, st. 8c// Sn. 41c, st. 9c//? 
Sn. 37b, st. 10c//? Sn. 36c), st. 1 labd// Sn. 37abd, st. 1 lc// Sn. 35c. and 
st. ll=st. 12 with jnati for putram in line c. This table is not quite 
complete, for there are many words in the two versions which are quite 
different in their corresponding lines. Stt. 6-10 are mere repetitions of 
the same idea with a different word in line c. in each stanza. In the 12 
stanzas of the Khadgavisana Gatha could be seen the theme of the Pali 
sutta fully discussed and developed, and likewise the seven corresponding 
stanzas in the Pali deal with the topic to a satisfactory degree. The rest 
of the stanzas express the same ideas in different words dwelling on the 
theme at length. 

There is an apparent contradiction in Sn. 45 when it mentions a nipakam 
sahdyam as contrasted with naputtam iccheyya kuto sahayam (Sn. 35c, cp. 
Sn. 37, 40 and 41). This kalyana mitta, as other texts would have it, is 
not to be categorised as a santhava, according to the sutta. The same 
idea is reflected at Sn. 94, 185, 187, 254 and 255; and Sn. 338 in Rahula 
Sutta makes specific mention of kalyana mitta. It is interesting to note 
that this topic is discussed at two different places in the sutta (viz. Sn. 
45-47 and Sn. 57-58). This shows that either the intervening stanzas 
were interpolated at a certain stage or Sn. 47 marks the end of the section 
dealing with mitta and that Sn. 57-58 were added later. (The concluding 
stanza too makes a casual reference to this type of ' 'noble companion ' '). 
The internal evidence of the sutta does not necessarily warrant such a 
conclusion if the criticism is based on linguistic data and other evidence 
alone. The sutta differentiates between two kinds of friends those in the 
household life; e.g. Sn. 40-41 and those in the brahma-cariya; e.g. Sn. 45, 
47, 50.- Perhaps it is possible that the "friends in brahma-cariya " is an 
allusion to the acariya-antevasika and upajjhaya-saddhiviharika relation- 

Pa// Buddhist Review 2, 1 (1977) 25 

ships in monastic life. The insistence on a life devoid of any associates 
was perhaps felt to be too exacting and therefore a compromise was 
reached by putting forward the "ideal companion" 13 A. K. Coomara- 
swamy (H. O. S. Miscellany of Pali Terms, s.v.) equates kalyanamitta 
to mahittma or mahatta; but this is not very convincing. The uniformity 
of the language of these stanzas and the absence of other evidence prevents 
one from classing some verses to be earlier or later than the rest. It may 
be only probable that the stanzas in Mvastu. preserve an older version, 
though both Pali and BSk. may be traced to an older source which is' 
now lost. 

It is also noticeable, from the repetitions in stanzas 6-10 and ll-12(in 
Mvastu.) that the version there is also an enlargement of an earlier sutra 
but it seems, on the whole, to represent an earlier stratum than the Pali, 
though the latter will be seen later to be considerably old. The possibility 
of the BSk. being a condensed version of an earlier sutra is out of the 
question for as a rule, no such tendency could be observed in BSk. works, 
and it is customary for them to contain expanded and enlarged versions 
of the same sections that are found more briefly in Pali. What is significant 
here is that the gathas in Mvastu. are far less enlarged than the correspond- 
ing sutta in Pali, and besides, the stanzas do not occur in the order in which 
the corresponding stanzas occur in Sn. A stanza parallel to Sn. 36 occurs 
at Divy. 294. It runs : 

Samsevamanasya bhavanti snehdh 
snehanvayam sambhavatiha duhkham/ 
adinavam snehagatam viditva 
ekas caret khadgavisanakalpah// 

(Attachments arise to him who associates with companions: misery in this 
world comes into being through attachment. Realizing the evil con- 
sequences bound up with attachments let him wander alone as the rhino- 
ceros. The stanza that bears the closest resemblance in Mvastu. is st. 10, 

Samsevamanasya siyati sneho 
snehanvayam duhkhamidam prabhotil 
putresu adinavam sammrsanto 
eko care khadgavisanakalpoj '/ 

The occurrence of this stanza in Divy. may equally suggest that both 
Mvastu. and Divy. have drawn from an original Khaggavisana Sutta which 
is perhaps preserved in entirety in Sn. along with subsequent additions and 
there is sufficient proof to show that the Pali version is an enlargement of 
an earlier existing nucleus. The fact that the Pali sutta abounds in lyrical 

13. I am indebted to Miss I. B. Horner for this observation. 

26 Pali Buddhist Review 2, 1 {1977) 

beauty and that its general diction of poetical expression is highly refined, 
the existence of a super-abundance of similes and the use of illustrative 
examples (e.g. Sn. 48) are in support of it. The uniformity of the stanzas 
in language, syntax, style and metre shows that the expansion has taken 
place very early. Both Nd2 and Ap. (I. 8-13) contain the Khaggavisana 
Sutta in full, and this shows that the sutta as is found now was known 
from comparatively early times. 


Before examining linguistic and other internal data it would be of some 
use to see how later writers looked upon this sutta. . The Corny, and Nd2 
divide it up into four vaggas. The division is as follows:— 

Corny. Vagga I, Sn. 35-44; II, Sn. 45-54; III, Sn. 55-64; IV, Sn. 65-75. 

Nd2. Vagga I, Sn. 35-44; II, Sn. 45-55; III, Sn. 56-65; IV, Sn. 66-75. 
The Commentator states that all the stanzas were uttered as udana by 
Pacceka Buddhas and gives the atthuppatti (context) of each stanza with 
the stories of these Pacceka Buddhas, some of whom he mentions by name. 
The 41 stanzas of the Khaggavisana Sutta mz incorporated in the Pacceka- 
buddhapadana {Ap. I, 7 ff.). The additional gathas there (i.e. 1-7 and 
50-58) serve as an introduction and a conclusion respectively. An extra 
stanza is added to the Khaggavisana Sutta proper, i.e. stanza 8 which 
differs from 9 (=S«. 35) only in line c; mettena cittena hitanukampi (= 
Mvastu. st. 2c). The Corny, of the Apadana too mentions the names of 
several Pacceka Buddhas, but they are different from those given in SnA. 
The inclusion of this sutta in Ap. and the fact that it is commented in Nd2 
prove that it was known to the compilers of these respective works as it 
exists to-day. The independent existence of this sutta prior to the com- 
pilation of Sn. is seen from Ndl and Mvastu. which do not place it in a 
particular group such as the Uraga Vagga. 


This sutta, like the Uraga Sutta, is undoubtedly meant for the benefit 
of the muni and belongs to that category of suttas which may be termed 
the "»ww'-class". Forty of the forty-one stanzas contain the refrain 
exhorting one to lead a life of solitude. 14 

14 Sn 45 which contains no refrain is to be found at Vin. I, 350, M. Ill, 154, Dh. 
328 329 3 III 488 and Dh.A. I, 52 along withSrc. 46. In the above instances the line 
eko care m'atan'garanne va nago (vide I) is to be seen in place of the usual refrain. It is 
probable that the simile with the elephant was earlier than that with the rhinoceros 
whose solitary habits were not so well-known as those of the elephant. It is significant 
that in the older "lists"of wild animals khagga is not mentioned. (/.V. 416 is obviously 
late) In view of the above facts it is highly probable that Sn. 45 and 46 were impor- 
tations to this sutta and that the line d of Sn. 46 was changed to suit the sutta. 

Pali Buddhist Review 2, 1 {1977) 


The language of the sutta, on the whole is rather old, and may be said 
to belong to stratum of early gatha-?&\\. Old forms, both verbal and 
nominal, archaic compounds, the vocabulary free from any late words, 
the simple constructions and very easy syntax suggest that the gathas are 
rather old. The easy and fluent style and the diction which is definitely 
poetic add much to the lyrical beauty of the poem. The abundance of 
similes and the occasional imagery used may lead one to assign a more 
recent date to the poem, but these two facts merely emphasise the merits 
of the sutta as a ballad. The absence of anything artificial or laboured 
removes all doubts of its early date. The external evidence from Nd2 
and Ap. is quite overwhelming in favour of a comparatively early date, 
though Mvastu. seems to suggest that there may have existed a version 
still earlier than that found at Sn., from which both Sn. and Mvastu. 
developed their respective versions. 

The metre of the poem is regular Tristubh with anacrusis 15 and jagati- 
padas 16 in a few lines. Neumann {Reden, p. 413) points out tmesis in Sn. 
53b, which should normally read, saitjatapadumikhandho ularo. Tmesis 
is a very old poetical device which is rather frequent even in the Rgveda. 

The sutta contains many linguistic forms that may be classified as old. 
There are three old ppr. forms in -aw, old absolutives as chetvana Sn. 44c, 
bhetva Sn. 62b, agent nouns like sahitd Sn. 42c and sammasita Sn. 69c, 
many historical absolutives ending in -ya, e.g. annaya, vineyya, Sn. 58c, 
abhibhuyya, Sn. 45c, etc., optative 3rd singulars in -etha, e.g. labhetha, Sn. 
45a, 46a, etc. (usually confined to the poetic language), probable dialectical 
forms as kammara- Sn. 48b, suhajje Sn. 37a, and poetical forms as seritam 
Sn. 39c, 40c, vaco (Fedic) Sn. 54c, rakkhitamanasano Sn. 63b, upekham Sn. 
67c, 73a, apekha Sn. 38b, and many elements which can be traced to Vedic, 
e.g. atho, etc. Some of the numerous cpds. used in the sutta seem to have 
become stereotyped already. Metrical lengthening is to be seen at Sn. 
38c vamsakaliro, Sn. 49a saha, Sn. 6 1c mutima and Sn. 70b Satima. Dukha 
is found for dukkha at Sn. 67a probably on the analogy of sukha or for 
purposes of metre. Similarly atthanam and karanatthaya are contracted 
to atfhana Sn. 54a and karanattha Sn. 75a respectively. Judging from 
these instances the sutta as a whole bears a stamp of antiquity. 

A few linguistic forms and other peculiarities of interest are: — Khagga- 
visanakappo Sn. 35d-44d and 46d-75d (already discussed), vide Ndl. 129 
and SnA. 65. This sutta abounds in cpds; some of them like yenicchakam 
Sn. 39b, itaritarena Sn. Alb, yathabhirantam Sn. 53c, analamkaritva Sn. 
59b, are of special interest here as they occur in the prose canonical idiom 

15. Vide Helmer Smith, SnA. 638. He points out anacrusis in Sn. 35b, 40c, 41c, 
45c, 59b, 63c, 68c, 69c, and 71c. 

16. ibid. Sn. 47a, 50a, 60ab, 66a and 70c. 


Pali Buddhist Review 2, 1 {1977) 

as well. Sneho Sn. 36a cp. 36c. Both sneha and sineha occur in this 
sutta: see sinehadosam at Sn. 66c. There is no hard and fast rule regarding 
the consonantal group sn- in poetry, though prose generally prefers 
the forms with svarabhakti; (also vide Geiger, §52). Statistics would 
throw hardly any light on this point, for the use of forms with or without 
svarabhakti is mainly governed by metrical exigencies and poetic idiosyn- 
crasies. Pahoti Sn. 36b is used in both prose and verse in the sense of 
"arise" though pabhavati is restricted to poetry (s. v., P.T.S.). Pekkha- 
mano Sn. 36c, etc. There are 18 medial ppr. forms in-mana in this sutta. 
Of the 350 ppr. forms in Sn. as many as 139 are medials, 107 of which 
end in -mana. The fact that this form is used in all periods of Pali does 
not preclude the possibilities of the stanzas being old when other corro- 
borative evidence is taken into account. Suhajje Sn. 2>1& (cp. kosajja) 
appears to be a dialectical word. The Pali word parallel to Sk. suhrd is 
suhada, but this form probably was an analogical derivation from the 
abstract sau-hrd-ya>sohajja. The other possibility is that sohajja the 
secondary form from suhada became suhajja by the weakening of the 
vowel o; o>u cp. Sk. &yaw>MagadhI aso>]?. asu also Gen. pi. gunnam 
(Sk. gonam) and Sk. sadyah>'P. sajju. Sabita Sn. 42c (cp. sammasita 
Sn. (69c). There are 21 such historical agent nouns in Sn. which should 
all be ascribed to an early stratum in Pali though canonical and later 
prose also contains them. Atho Sn. 43b, atho is formed from the copul- 
ative (and adverbial) particle atha and the enclitic u, and can be traced 
back to the later hymns of the Rgveda and the Sathapatha Brahmana 11 : 
This compound particle appears to be restricted to poetry and occurs 
no less than 25 times in Sn. alone. Saddhimcaram Sn. 45b, 46b, saddhim 
— cara (the adjectival suflix from y/car) cp. dada in pannadada Kh. VIII 
10 or kamadada Pv. II, 9.1. As the cpd. is formed from the indeclinable 
saddhim and it retains the nasal as in analamkaritva Sn. 59b, rattimdivam 
Sn. 507c, 114b it is of special interest. KammSra Sn. 48a is a Prakritism 
used in all stages of the language, in the specialised meaning of "smith". 
Sk. karma-kara>J > . Itamma-kara>'kamma-ara (cp. ajja-utta for urya- 
putra)>P. kammara; cp. Krsinagara>P. Kusinara. Phassaye Sn. 54b 
is probably a dialectical form. The root sprs is treated as a verb in class 
X, perhaps on the analogy of forms like cintaye. The direct historical 
forms should be phasse and phuse. RakkhitamanasanoS/?. 63b. notn. sg. 
is formed by adding the adjectival six. -na to manasa the secondary form 
of manas. This too is a pure poetic form. 


The doctrinal import of this sutta has already been touched upon. 
It has been emphasised earlier that the quest of the secluded life pertains to 
17. Vide MacdonelJ, Vedic Grammar for Students, pp. 214-215. 

Pali Buddhist Review 2, 1 (1977) 29 

the earliest stage of Buddhism and sheds much light on the life of the 
hermits (munayo). A noticeable development in doctrine in the sutta 
is the concept of a noble companion (39). It has been pointed out above 
that the Khadgavisana Gatha in Mvastu. make no mention of this type of 
companion. If the version in Mvastu. is considered as representing an 
earlier form of this sutta, perhaps a form nearer the nucleus out of which 
the present long sutta has developed, it may be justifiable to infer that this 
concept is a later accretion. On the other hand, it is more probable that 
the idea of a kalyana-mitta developed in the Theravada School before the 
time of composition of the Pali Khaggavisana Sutta. The references to 
kalyana-mitta (virtuous companion) at Sn. 338a, mittasampadam (good 
companionship) at Kh. VIII, 14, sahaya-sampadam at Sn. 47a, etc. (s.v., 
P.T.S. for more references) do not make it clear whether the idea develop- 
ed early or not, but the idea of the kalyana-mitta as the spiritual advisor 
or guide appears frequently in younger contexts (s.v., P.T.S.). The term 
parallel to the earlier concept in Buddhism is to be found in sddhusanga 
of the epics (Mbh.). It is not in the latter developed meaning that these 
terms occur in this sutta. Although this idea is rather contradictory to 
that of ekacariya it cannot be considered as late. The uniformity in 
language and metre makes the possibilities of a wholesale interpolation 
improbable. The lack of consistency in the logical trend of the sutta may 
indicate that the verses had existed earlier in some unsettled order and 
that the present order is due to the efforts of a monastic editor. 

Another important concept that is taken for granted is metta. It is 
alluded to at Sn. 42a, Catuddiso appafigho ca hoti. (He has no conflicts 
from the four quarters), and is mentioned later on in the sutta at Sn. 73 
along with upe(k)kha, karuna, vimutti and mudita. The idea of metta 
(friendliness, amity) is a central concept in Buddhism, both early and late. 
Four of the five items mentioned at Sn. 73 came later to be classified as the 
brahmaviharas. Besides the fact that no specific mention of the brahma- 
viharas is made, the four items which constitute it do not occur here in 
their classified order; i.e. metta, karuna, mudita and upekkha. There is 
no doubt whatsoever that these concepts go back to the earliest phase of 
Buddhism and perhaps Mrs. Rhys Davids is right when she speaks of 
brahmavihara as a later term for these four items, 18 though the name itself 
is not late and is applied to metta alone at Sn. 15 Id. This sutta thus 
reflects a time prior to these concepts being labelled as brahmaviharas. 
The expression anhaya atthani at Sn. 58c (having known the atthas) 
demands attention. The explanation at Nd2, 85, atta'ttha, para'ttha, 
ubhaya \ttha, ditthadhamma'ttha, samparayika' ttha and parama'ttha (own 

18. Mrs. Rhys Davids, Outlines of Buddhism, pp 32 ff. and What was the Original 
Gospel in Buddhisml pp. 92 ff. 


Pali Buddhist Review 2, 1 (1977) 

welfare, others' welfare, welfare of both, welfare in this world, welfare 
after-death and highest welfare), merely suggests the various implications. 
SnA. Ill agrees with this explanation. It is quite probable that attha 
here meant not only paramattha — the summum bonum in Buddhism, but 
embraces a still wider meaning as suggested by the commentaries and is 
probably connected with the attha suggested earlier in connection with 
the Atthaka Vagga (PBR 1, 3, p. 143). 

All the other terms and topics of doctrinal import in the sutta are to be 
met with in other Pali works, both old and young, and therefore demand 
no particular attention. Worldly attachments and ties (Sn. 35 if.), lustful 
tendencies (Sn. 50), materialistic leanings (Sn. 54), and perverse views 
(Sn. 55), are denounced. The five obstacles to progress (mentioned by 
number only) are to be abandoned (Sn. 66) and upe(k)kha is to he develop- 
ed (Sn. 67). The positive side of the life of a muni discussed in Sn. 65-74 
necessitates the mention of many terms which have acquired a technical 
significance. The complete list of instructions at Sn. 69 may seem to . 
appear rather late on account of the fact that many important concepts 
are heaped together, but the haphazard manner in which the items occur 
does not show any sign of lateness. Moreover, all the topics mentioned 
there are quite consistent with the general theme of the sutta as well as the 
life of the early hermits. Rdga, dosa and moha which are mentioned at 
Sn. 74a suggest that they have almost reached the stage of being classified 
into the stereotyped group of the three akusalamulani; but the term as such 
does not occur here. Generally speaking, the sutta on doctrinal evidence 
represents an early phase of Buddhism. 


Other internal evidence consists of social conditions reflected in the 
Sutta and other casual references. As far as social conditions go not 
many data can be gathered, as the sutta paints a picture of the life of 
recluses only. The reference made to some (eke) discontented pabbajitas 
at Sn. 43 may be an allusion to a contemporary sect or class. It is difficult 
to say who these recluses were from the scanty evidence available. The 
stanza seems to bear a faint connection with Sn. 45ab, which can be 
considered as referring to the philosophy of such a sect. Yet, it is not 
possible to establish a definite link between the two, as samayikam vimuttim 
may not refer to any particular view, but to temporal joys. 19 It is only 

19. Vide SnA. 105, samayikani vimuttin ti lokiyasamapattirn, sa hi appitappitasamaye 
eva paccamkehi vimuccanato samayika vimutti ti vuccali (cp. PtsA. Ill, 552 ff.) — Temporal 
emancipation means worldly attainments. It is called temporal emancipation because 
whenever one indulges (in these pleasures, cp. Sn. 54b) one is emancipated from what is 

Pali Buddhist Review 2, 1 (1977) 31 

probable that these two stanzas refer to a sect of materialists (Carvakas). 
There are numerous references to materialists and their doctrines in the 
Nikayas (Samannaphala Sutta, etc.), and according to Rhys Davids, they 
must have preceded Buddhism as early Buddhist literature mentions 
them. 20 

The line Sn. 75b, nikkarana dullabha ajja mitta (friends without a motive 
are rare today) seems to refer to the time of composition of the sutta. 
This by itself is of no great significance, for human nature has been the 
same through the ages. Along with this may be compared Th. I, 
949-980 where Phussa prophesies that the future of the Sangha would 
be gloomy. The passage is a condemnation of the white-robed ascetics, 
and shows the rivalry between the ascetics and the monks. The prophecy 
actually alludes to the time of compilation of these gdthas. In the Sutta 
Nipata the significant point is the mention of the word ajja, though it 
does not in any way help to determine the date of the sutta. 


Taking all the evidence into consideration an early date may be assigned 
to the sutta. Linguistically, it is seen to preserve an early stratum of Pali. 
Doctrinally, it represents an early phase of Buddhism, tinged with the 
germs of some important tenets of that phase of Buddhism which came to 
be termed Theravada. External evidence within the Pali Canon itself 
suggests an early date for all the stanzas of the sutta, but evidence from 
Mvastu. and Divy. seem to indicate that the Pali sutta was an enlargement 
of an earlier nucleus. Metre shows that all the stanzas in the poem 
should belong to the same period if not to one author. The style too is 
uniform throughout the sutta. 



The Muni Sutta portrays certain characteristics of the muni— the sage. 
The poem agrees in theme with the Uraga and Khaggavisana Suttas. 
These three suttas together with Moneyya Sutta (i.e. Ndlaka excluding its 
vatthugatha) can be considered as the proper "Muni— ballads", though 
there is constant mention of the attributes of the muni in the greater part 
of the Sabhiya Sutta and many suttas of the Afthaka Vagga such as Jara, 
Tissametteyya and Magandiya. It has already been noted that the 
Uraga Sutta resembles this sutta in many respects. While the Uraga Sutta 
describes the ideal bhikkhu, the Muni Sutta gives a descriptive definition 


Pali Buddhist Review 2, 1 (1977) 

of the muni. Generally speaking, there appears no fundamental difference 
between the muni and the bhikkhu in early Buddhism, and the terms are 
interchangeable, except when muni specifically refers to the Buddha. 
The qualities attributed to the muni are often associated with the bhikkhu, 
and sometimes with brahmana, khattiya, vedagu and sottiya in a strict 
Buddhist sense. 21 Although bhikkhu and muni are virtually synonymous 
there seems to be a subtle difference between the two. While renunciation, 
pabbajja (becoming a religious mendicant), detachment and ekacariya 
(life of solitude) are emphasised of the bhikkhu, the muni is described as a 
person who plays a more important rdle. This is quite evident from his 
description in the Muni and Moneyya Suttas, and the type of epithets 
used about him. In addition to the possession of all the characteristics 
of the bhikkhu, there appears something nobler and more positive about 
him than about the bhikkhu. He is a more evolved being (bhdvitatta) 
who has reached spiritual attainments and instructs others as well. The 
term muni in Sn. is used in a much wider meaning than bhikkhu in Th. 1. 

As regards the epithets, the muni is called a mahesi (Sn. 208 d ) and is 
described as tadi(Sn. 21 9 h ),yatatta (Sn. 216 b ), sannatatta (Sn.. 216 a ) and 
thitatta(Sn. 215^). Besides the eight references 22 where Buddha is called 
mahesi, the true brahmana (in the strict Buddhist sense) is spoken of as 
mahesi at Sn. 646, b also the khinasava is referred to as a mahesi at Sn. 82 a 
and 48 l a , though the allusion is to the Buddha. The epithet tadi is 
rather puzzling as it cannot be easily differentiated from tadi (Sk. tadrk 
also P. tadiso) the demonstrative adjective. Yet, there are sufficient 
instances in Sn. itself where tadi is clearly used in the pregnant sense of 
ecce homo. The muni is ubhayeneva tadi at Sn. 712° (unchanged under 
both circumstances) in the Nalaka Sutta. Paramatthaka speaks of the 
bhikkhu as, pdramgato napacceti tadi (gone over yonder such — a steadfast 
one — is he who returns not; — Sn. 803 d ). The Buddha is called asitam and 
tadim at Sn. 95T, cp. Sn. 21 9 C asitam anasavam). The maggajina is called tadi 
at Sn. 86 d in the Cunda Sutta. The brahmana, khettajina (cp. Sk. ksetrajna), 
vedagu and sottiya — all of them in a Buddhist sense — are called tadi in the 
Sabhiya Sutta (Sn. 519-532). Another attribute of the muni — yatatta 
(self restrained) — is repeated at Sn. 723 a in the Moneyya Sutta. Homeless 
recluses are called yatatta at Sn. 490 b . It is practically the same idea 
conveyed by the term sannatatta (self-subdued). The brahmins of old 
are referred to as sannatatta at Sn. 284 b (Brahmanadhammika Sutta), and 
susahnatatta occurs at Sn. 464 b (Sundarikabharadvaja Sutta). The muni 
is known to be thitatta (steadfast), so also is the virtuous monk described 
in the Kirnsila Sutta (Sn. 328 d ) the bhikkhu who renounces the world in 

21. See Sabhiya Sutta Sn. Ill, 6. 

22. Sn. 176<J, 177d, 9J5b, 1054s 1057*, 1061b, 1067b, and 1083a. 

Pali Buddhist Review 2, 1 (1977) 33 

the proper manner (Sn. 370 c ) and the good brahmana (Sn. 51 9 b ). Further, 
it is used as an attribute of the Tathagata at Sn. All. b The muni is also 
called asitam and anasavam (independent and free from the banes). The 
Buddha is described as asito at Sn. 25 l d and 957\ The true brahmana 
(brahma in the text) is called asito at Sn. 519 d . Again the monk who has 
drawn out the dart is described as asito at Sn. 593 a and so is the muni in 
the Moneyya Sutta. The manava Dhotaka begs for instruction so that 
he may lead a life of peace and independence; (idheva santo asito careyyam 
—Sn 1065 b ). 

It is evident that these standard epithets of the muni definitely speak of 
the positive side of his life. Many of these terms- are not employed to 
describe the bhikkhu though he may possess the qualities which these 
epithets attribute to the muni. There is some implied superiority of the 
muni over the bhikkhu though the ideal of the bhikkhu is in no way to be 
understood as falling short of that of the muni. 

There are various other attributes of the muni enumerated in the sutta. 
He has no fixed abode and he is free from any acquaintanceship (Sn. 207). 
He has eradicated all sin; and is the lonely wanderer (cp. Sn. 35-75) who 
has visioned the state of peace (Sn. 208). He sees the ultimate destruction 
of birth, leaves reasoning behind and is under no limitations of time and 
space (Sn. 209). He is free from covetousness and has reached the 
Beyond— para— (Sn. 210). He is sabbabhibhu— one who overcomes every 
obstacle and is superior to all others;— he has perfect knowledge and is 
unsmeared by the worldly phenomena and is emancipated. (These are 
the attributes of the Bhagava mentioned at M. 1, 171 S. II. 284, Vin. I. 
8 and Dh. 353)— (Sto. 211). He is wise and composed, and is free from 
the mental obsessions; he delights in meditation, wanders alone and leads 
others (Sn. 212-213). He is firm and straight, discerning, free from lust 
and he shrinks from sin. His senses are serene and he is endowed with 
propriety of speech (Sn. 214-215). He is self-restrained and self-subdued 
(Sn. 216). He knows the world and sees the highest attha (well-being). 
He has crossed the ogha (flood) and the samudda (ocean), has cut off all 
knots, has nothing to lean on and is steadfast. 

The sutta in every respect is Buddhistic and the terms and values in it 
bear ample testimony to that effect. The simple allegory taken from the 
uprooting of a tree or of not sowing the seed (of tanha) developed to a 
considerable extent in Sn. 208-209, the shunning of resting places of the 
mind (nivesanani) — Sn. 210, cp. also Atfhaka Vagga, remaining unsullied 
by worldly phenomena, and such other central concepts of early Buddhism, 
prove that the sutta contains very early Buddhist sayings and there is much 
evidence to show that the subject-matter of this sutta is very old. The 


Pali Buddhist Review 2, 1 (1977) 

Pali Buddhist Review 2, 1 (1977) 


theme of the sutta (as well as that of many other poems of Sn.) is closely 
connected with the yogic ideal. The muni chooses with stoic indifference 
the middle way between self-mortification and attachment to worldly 
enjoyment. This ideal as current in pre-Asokan times coincides with the 
yogic ideal promulgated in the Gitd (Bhg. II- VI. cp. Bhg. II, 56; II, 69; 
V, 6; V, 28; VI, 3, etc., where the muni is mentioned in similar words). 


The stanzas themselves need a close examination before the internal, 
external and indirect evidence is taken up for discussion, A simple 
analysis of the sutta shows that its stanzas fall into three groups; viz. — 

Group 1. Sn. 207-210, 
Group II. Sn. 211-219 and 
Group III. Sn. 220-221. 

Group I. — The four opening stanzas seem to form an independent unit 
— a short poem by itself.- Unlike the nine stanzas that follow, these verses 
do not contain the refrain (tarn vdpi dhira munim vedayanti); but it is quite 
significant that the word muni occurs at least once in every stanza of the 
whole sutta. Sn. 207 furnishes the introduction to the independent unit 
as well as to the whole poem. A noteworthy feature of this stanza is that 
it is in Anustubh sloka whereas the rest of the poem is in Tristubh. The 
stanza itself cannot, on this account, be brushed aside as a late introductory 
verse, for it was obviously known to have belonged to the Muni Sutta 
at least some time prior to the compilation of the Milindapanha. 25 The 
stanza itself breathes the same tone as the opening verses of the Khagga- 
visana Sutta — cp. etam bhayam santhave pekkhamano, Sn. 37 a . Game 
akubbam muni santhavani at Sn. 844 b is also reminiscent of the opening 
pada. The same idea is expanded at Dh. 212-216, viz. — Dh. 212 Piyato 
jayati soko piyato jdyati bhayam (from what is pleasant arise grief and fear) 
— , Dh. 213 pemato. .(from affection. .), Dh. 214 ratiya. . (from lust.'.), 
Dh. 215 kamato. .(from sense-pleasures. .) and Dh. 216 tanhaya. .(from 
avarice ..). All these causes of suffering or sorrow seem to be embraced 
by the term santhava, cp. also /. IV. 312. 

The next stanza (Sn. 208) introduces the familiar Buddhist allegory 

(already referred to) in which santhava (Corny, tanhd) is the tree that has 

to be uprooted. What has arisen has to be annihilated (eradicated), it 

should not be allowed to grow anew (lit. not replanted), and it should not 

be allowed to grow up when it has sprung. This allegory is further 

23. Milp3&5 quotes this stanza thus: — Bhasitam-p'ctam-maharaja Bhagavata deva- 
tidevena Suttanipate, and quotes Sn. 207. The stanza is the topic of a panha at Milp. 
212 ff. and is quoted several times there. 



worked out in the next stanza (Sn. 209). The vatthiini — lit. fields or bases — 
have to be reckoned, and the seed (Corny, abhisankharavihhana — ' 'storing 
intellect") has to be destroyed (Corny, himsitva, vadhitvd — Sk. pramdrya 
fromVmr. mrndti) — and it is not to be watered with sineha (desire). Sn. 
210 forms the conclusion of this independent unit. Judging by the ideas 
in them these four stanzas, taken by themselves, appear to be very old. 
This is further strengthened by the Commentator's testimony. He says 
that the uppatti (origin) of the whole sutta is not the same (Sn.A. 254). 
He gives the same uppatti for these four stanzas, but gives separate uppattis 
for each succeeding stanza. The Commentator's introduction seems 
rather strained and reports a somewhat incredible incident found also at 
A. III. 67-69 (Mataputtika Sutta // o A. III. 559). This tradition though 
as late as cc. 5th century a.c. cannot be totally ignored, as it is supported 
by the Anguttara Nikaya. On the other hand even if there is no connec- 
tion between the incident narrated in Sn.A. 254 ff. and these four stanzas, 
the very fact of the existence of the strong tradition that these four stanzas 
were found together, the internal evidence of the subject matter, and the 
recurrence of the opening stanza four times in Milp. attest to their great 

Group II. — The refrain is found in all these stanzas and all of them are 
uniform except Sn. 213 which contains seven padas instead of four. The 
three additional padas are the same as Sn. 71 abc in the Khaggavisana Sutta. 
The influence of the Khaggavisana Sutta is felt in ekam carantam munim 
appamattam at Sn. 21 3 a and, tarn ahu ekam muninam carantam at Sn. 208°, 
besides the repeated padas Sn. 213 cde . The four lines would have been 
complete and the stanza would have passed without special notice, but 
for these additional padas which in all probability were interpolated later. 
It is very unlikely that the whole stanza was an interpolation, although the 
ideas contained in it are closely related to the Khaggavisana Sutta. Similarly 
Sn. 211 cannot be considered as an interpolation though the first three 
padas occur elsewhere in connection with the Buddha 's meeting with the 
ajivaka Upaka. 24 In all these instances these words are put into the 
mouth of the Bhagava making him utter a boastful statement, which is 
quite contrary to his usual reticence about himself. It is quite probable 
that this was the original occurrence of these lines and that other texts 
may have drawn upon this stanza in reporting the incident between the 
Buddha and the mendicant Upaka. The repetition of the simile, tasaram 
va ujjum (like a shuttle that is straight) at Sn. 464 b and 497 b does not 
indicate that the idea has been borrowed in any of these instances, 
but that it was the common property of the poetical language. 

24. Vide Ariyapariyesana Sutta, Vinaya Introduction, etc. 


Pali Buddhist Review 2, 1 (1977) 

These nine stanzas form, as it were, the body of the Muni Sutta. They 
constitute the ballad proper, with the emphasis on the muni clearly brought 
out by the refrain and the similes at Sn. 213 cde , 214 a and 21 5 a . These 
stanzas form a homogeneous unit, just as the first four stanzas form a unit by 
themselves. The chief ideas of this section are seen to occur again in 
other ballads of Sn. as well as in other metrical works. Padas and lines 
of many stanzas are also repeated in other metrical works. 25 

According to Buddhaghosa the 1 1 stanzas beginning with Sn. 211 were 
uttered on various occasions by the Buddha and these sayings were 
gathered from various isolated instances and knitted together into a 
composite sutta. The uppattis (origins) given by him are: — 

Sn. 211 — uttered on the occasion of Buddha's meeting the ajivaka 
Upaka. (SnA. 258) cp. Corny, on Dh. 353. Sn. 212— uttered about 
Khadiravaniya Revata. (Sn.A. 261) cp. Corny, on Dh. 98 and Dh. 412, 
Sn. 213— preached to Suddhodana. (SnA. 262). Sn. 214— uttered after 
Cincamanavika's attempt to malign the Buddha. (SnA. 263) cp. Corny, 
on Dh. 176 and Sn. 780. Sn. 215— preached to the girl who was inspired 
by the straightness of the movement of the shuttle. (SnA. 265). Sn. 
216— preached on the occasion of the weaver girl's solution of the Buddha's 
riddle. (SnA. 266) cp. Corny, on Dh. 174. Sn. 217— preached to the 
Paftcaggadayaka-brahmana. (SnA. 270) cp. Corny, on Dh. 367. Sn. 
218 — preached to the monks, announcing the attainment of arahatship 
of a monk who had fluctuated four times between home and homelessness. 
(SnA. 272). Sn. 219 — preached to the monks, announcing Nanda's 
attainment of arahatship. (SnA. 273) cp. Corny, on Dh. 15. Sn. 220 — 
preached to the monks, announcing the arahatship of a monk who was 
alleged to have aided a hunter. (SnA. 275). Sn. 221 — preached on the 
occasion when the Sakiyas argued that a Sotapanna, even if he is a house- 
holder, should be honoured by another who reaches that stage subsequent 
to him. (Sn.A. 276). 

It is not at all probable that these stanzas were ' 'independent utterances" 
made on ' 'various occasions ' ' as Bdhgh. says. The coherence of thought 
and the inter-dependence between the succeeding verses and those preced- 
ing, indicate to what extent these stanzas are connected with one another. 
In all probability these nine stanzas (and perhaps Sn. 207-210) were the 
work of a single poet though it is very doubtful whether the two concluding 
stanzas too belonged to him. The significance of the diversity of the 
uppattis of these stanzas given by Bdhgh. is that the verses themselves 
were so well-known that there were separate stories appended to them 
by Commentarial tradition. This perhaps speaks of the popularity that 
these verses enjoyed. 

25. Vide E. M. Hare: Woven Cadences, p. 190. 

Pali Buddhist Review 2, 1 (1977) 


Group in —The last two stanzas of the poem stand out as a class by 
themselves as (a) they contain no refrain (b) point out the difference 
between the householder and the muni. Although these two stanzas 
are similar to the rest of the poem in language, metre and style, the change 
of values and the contrast made between the householder and the mum 
(emphasising the latter's superiority in the simile at Sn. 221* ) are indicative 
of an effort made by a monastic editor to usher in to this poem of lofty 
theme and ideals, an idea of comparative insignificance. Their position 
as the last two stanzas of the final sutta (of the vagga) suggests that they 
may as well have been intended to form the very conclusion of the whole 
vaLa In view of their decidedly late characteristics and the indirect 
evidence from the position of the sutta it is clear that these two stanzas 
were added when the sutta was re-edited before its inclusion in Sutta 
Nipata. 26 


The language of the sutta is old-and all the stanzas are composed in the 
archaic poetic , dialect often called "g^a-Pali". There are many 
linguistic forms in the sutta that belong to an early stratum of Pa i, and 
forms which may be classed as late are totally absent. The usual historical 
verbal forms like Optative 3 sg. P. pada m-e, A pada m-etha the 
historical sibilant Aorist (addakkhi, Sn. 208 b ), absolutives ; in -ya (7 of 
them), old present participles in-am (anikamayam Sn. 210 ) and old 
infinitives (like thutum Sn. 217 c ) as well as nominal forms such as rajo 
(Sn. 207"), muninam (Sn. 208 c ), gedha (Sn. 210°), and ubhc >(Sn 220 ), 
dialectal or local forms such as anuppavecche (Sn. 208\ 209 b ) and ujjurn 
(S».215*),and historical particles like ve(Sn. 207*; Vedic v«0etc. show tha 
the language of the sutta is old. The linguistic forms of interest « jtto 
sutta are-.-ropayeyya, Sn. 208*; The causative is formed after verbs of 
class X, and -eyy- is added. This is a pure Pali form. Assa ,Sn. 208 
This should be interpreted as a dative, as in Corny. U.. as m*«""« 
with elipsis-(S^. 256). The two lines would then read:-Yo jatam 
ucchijjajayantam na ropayeyya, assa (jayantassa) ™"^ avec ff. J"J. 
pavecche, Sn. 208\ 209 h . (Corny, anuppaveseyya, ™™Maneyya 
!hould permit entry or fulfilment). There are three possible explanations 
to this curious form:-l-(Trenckner). ^^yaccha> yeccha->veccha- 
%y> v in ayudha>avudhaet^i^^ 

-^n^T^charac^ diideto 

recensions before it took its final shape in Sn. At present '/'* n ° £. ibility of the name 
what recension of it the title M«^^P fSJjES by Max ^alleser, also 
referring to all poems in praise of the «««-««» « l„ irrelatively late as compared 
needs consideration). The inclusion of the sutta in in. is ro« 
with the date of its composition. 


Pali Buddhist Review 2, 1 (1977) 

-a- >-c- cp. sa-yatha>seyyatha, {pavekkhe etc.) 2-V Wi>Caus. vesyet 

( f }> :T~>~ CChe - CP - - tS - > - SS - H- cch - ( Sk - utsa *Sa. P. ussanga II 
ucchanga).™. (If lt could be established that -i->-cch, just as s-> ch- 
(mmally) as m chakana, chava, chapa or cheppd, the form might as well 
bedewed from opt. of Vvtf, i.e. *vesyet on the analogy of class IV verbs) 
3-The optative of the future base of V vis i.e. *vek-> *vekkhe (// E. Pkr ) 

L ftf t0 ^ Pkr ° W ° Ul4 be thC Same aS this form - 28 Thambho-r- 
f lu I '■ , 1S an lnor 8 anic san dhi with the artificial replacement 
of the historical-A which, with the preceding a i.e. -aft, has already become 
-o;iva is retained as in Sk. Ubho, Sn, 220% <u bhau, the Vedic dual 
It is a historical form. Ujjun,, Sn. 215% dialectal or Pruritic ,» 
*/«-f./« (with epenthesis)> l(/ yK. Naiam thutum, S«. 217 c - the old 
C thutm Cii ° n Wlth "^ 3nd thC infinitive; Sk - stautum>*thotum->V. 

Metre -The difference in metre between the opening stanza and the 
res of the poem has already been commented on. This difference of 
metre in the introductory verse is quite similar to that in the opening verse 
m Kimnla Sutta. It was probably on verses like these that the practice 
rJZ?7 g f 'f with vatthugdtha was based. The opening stanza is a 
regular Anustubh sloka with an odd quarter in line c. The rest of the 
poem is in Tnstubh metre with Jagati padas at Sn. 212° 2l4 ac 218 b 219° 

Im^Z'^l'rr^^ aftCr thC ?th SyllabIe is not reckoned i>n ^. 
a r , ( 7fl » aft ). There are two difficult padas (Sn. 214 a and 215 b ) 

and Helmer Smith (Sn.A. 639) suggests ^«„e and kammahi for ^ 
vgahmetndkammehi respectively. There is metrical lengthening in 
santhavato (Sn. 20T), muni (Sn. 209', 210% 216% and 220 d v 1 mum) 

Tn^ f^r 2W ^h mUPalittam ^ 21lb V ' L ■"") and satimam^n. 
212 ). Both forms gthi (Sn. 220 a v.l. -I) andgihi(Sn. 220 c and 221= v 1 -A 

e°mnToveH?n 0, ;f metre ' f aS a ru,e ' is no sound criterion, the old metres 

tte p7em is old '"" a f SUPP ° rt ° ther eVidenCe WbJch sh ° WS that 

Th^lL^ ^ n ° thin ! extraordinar y as re gards the style of the sutta. 
Throughout the poem the diction of the stanzas is purely poetic and 
figurative speech is freely used, e.g. Sn. 209% 219 b < etc. Simile and 
metaphor play an important part. I n addition to the popular similes 
men icned already there is a simile at Sn. 221- The allegory Tin. 
208-209 also enhances the poetic effect. E. M. Hare (p. 218) considers 
ypamaya (Sn. 209*) is a slesa (pun) i.e. from V ^ andvW There 
are a few instances of alliteration (Sn. 211) and assonance (Sn. 219 b - 
atitanyaja dim). These poe dcal_device s which are n ot too frequently 

27. Vide M Oiler; Pali Grammar, p. 120 

28. Vide Geiger, § 152 note 3. 

Pali Buddhist Review 2, 1 (1977) 


employed in the sutta do not in, any way mar the style as in late artificial 
poetry. The refrain in Sn. 211-219, the popular similes used, and the 
perfect rhythm and cadence, all point to a literary style which is essentially 
that of the ballad, and therefore popular. 

Doctrinal Developments — Besides the epithets used for the muni, 
which have definite values and an underlying technical significance there 
are a few terms in the sutta which show a transition from the general 
meaning to a more restricted connotation indicating a gradual develop- 
ment in doctrine. The word santhava (Sn. 207 a ) implies not only acquaint- 
ances but also all forms of ties, attachments and worldly bonds arising 
out of the association with them (cp. Khaggavisana Sutta). The words 
vatthuni, bljam andsineham, though allegorically used have a faint technical 
significance; and these words in course of time came to be looked upon 
as synonyms for the various objects they stood for. This process is easily 
discernible, in the case of sineha on account of the semantic development 
of the word (sineha=\iquid and affection), and became most pronounced 
in the Commentarial epoch. Neumann 29 suggests that takkam and 
sankham at Sn. 209 d were references to popular philosophical systems 
Tarkyam and Sankhyam. It is not at all likely that takka and sankha 
were references to any definite philosophical system so much as to any 
speculative doctrine which professed to achieve salvation by way of 
knowledge— jnana as opposed to yoga. 10 Sankham cannot be an allusion 
to the philosophical system known as Sarikhya originated by Kapila, 
before Buddhism, but developed centuries later. It has been rightly 
pointed out by Neumann (ibid) that the muni seeks no resting place as the 
yogi as stated at Mbh. Santiparvan 302, pratyaksahetavo yogah, sankhyah 
sastravinicayah. The phrase sankham na upeti, however, is intrinsically 
connected with na sankham gacchati (does not enter the category of, or, 
is not reckoned as) occurring often in canonical Pali. The only point 
worth investigating here is to what category (lit. number) the muni does 
not belong. The explanation of this phrase in the Corny. (SnA. 257) 
that the muni does not enter the category of "a divine being or an(ordinary ) 
man ' ' or even of ' 'a person of lustful temperament or of malicious tem- 
perament" sheds some light. In short, the idea implied is that the muni 
is beyond worldly limitations — an idea quite in harmony with the con- 
ception of a perfect muni. 

The term nivesana (Sn. 210 a ) as 'a resting place for the mind, a dogma ' 
is a word adapted by early Buddhism giving it a special meaning. It has 
no special doctrinal significance, apart from the fact that this specialised 

29. Die Reden Cotamo Buddhos, p. 437. 

30. Vide Franklin Edgerton: "Samkhya and Yoga in the Epics "—American Journal 
of Philology, 1924. 

Pali Buddhist Review 2, 1 (1977) 



Pali Buddhist Review 2, 1 (1977) 

meaning was attached to it from very early times. The idea of being 
free from such nivesanani (ov-nivesa 785 a ) occurs also at Sn. 470 ab , 801 c , 
846 d , and 1055. c Sn. 785 describes the nature of dogmatic beliefs. The 
term para and the idea of 'going beyond '(para+Vgam) — Sn. 210 d have 
been discussed in the introduction to the Parayana Vagga. 11 The words 
ogha and samudda Sn. 219 b are used to signify the ills of the world in 
much the same way as vatthuni, bijani and sineha, but ogha seems to have 
already acquired a technical significance as seen from its occurrences 
in Sn. i2 

The doctrinal emphasis of this sutta is on the conduct of the muni. 
This itself shows that the sutta reflects an early period. Most of the 
terminology of the sutta, apart from the basic concepts such as sahga, 
santhava, etc. is not fixed. The terms used in the allegory of the seed and 
that of reaching the further shore of the samudda (ocean) are seen gradually 
to acquire a technical significance. This_ sutta furnishes a great deal of 
data for the development of early Buddhist terminology. All the available 
evidence from doctrinal grounds too shows that the sutta appears early. 


External evidence. — References made to the Muni Sutta in other works 
show that the sutta was known before the compilation of these respective 
works. The Muni-gatha are mentioned as one of the sections recited by 
Srona Kotikarna at Divy. 20, and by the merchants at Divy. 35. The 
Vinaya of the Mulasarvastivadins and the Tibetan Dulva include the Muni- 
gatha among the sections recited by Srona. 33 As Rhys Davids 34 has 
pointed out, by Muni-gatha was meant the Muni Sutta. In one of the 
places where Milp. (i.e. p. 385) quotes the opening stanza of the Muni 
Sutta (i.e. Sn. 207) the name of the sutta itself is not mentioned, though 
reference is made to the Sutta Nipata, (see note 6). Usually the author 
of Milp. refers to the whole work rather than to a particular sutta when he 
makes his quotations; e.g. Samyuttanikayavare, Suttanipate, etc. Al- 
together he makes five references to Sn. in quoting stanzas taken from it. 35 
There are other quotations from Sn. with no references to it whatsoever, 
and at one place (Milp. 36) he quotes Sn. 184 and acknowledges it as a 
stanza of the Samyuttanikaya (S.I. 214). It is only in one instance (Milp. 
369) that a sutta in Sn. is mentioned by name; viz. in quoting Sn. 29 he 
says, Vuttani' p'etam Maharaja, Bhagavata devdtidevena Suttanipate 

31. PBR 1, 3, p. 146. 

32. ibid. See also the introduction to the Parayana Vagga. 

33. Sylvan Levi, J.A., 1915, p. 401 ff. 

34. Rhys Davids. J.P.T.S., 1896, p. 95. 

35. viz. Milp. 369, 385, 411, 413-414, and 414. 

Dhaniyagopalakasutte. Now, the only occurrence of Sn. 207 in the whole 
work is as the opening verse of the Muni Sutta, and it may be said with 
certainty that the author of Milp. knew the Muni Sutta as belonging to 
Sn. Although Milp. is a comparatively late work(cc. 80 B.C.), 36 all these 
quotations show that Sn. was perhaps known to its authoras it is found 

The earlier inscriptional evidence from the Bhabru Minor Rock Edict 
of Asoka shows that the Muni Sutta was a popular piece even as early as 
the third century B.C. The fact that the Pali versions of the episode of 
Sona do not include the Muni Sutta among the pieces recited by him does 
not in any way prove that the sutta was not known to the compilers of 
these respective works. It is only in the more enlarged versions of the 
episode that the Muni Sutta as well as other well known sections of the 
Scriptures are mentioned. However, the testimony of the Bhabru edict 
is sufficient to show that the sutta was known in comparatively early times. 
The indirect evidence from the position of the sutta in the vagga has 
been discussed earlier. Yet, it should be observed that the inclusion of 
the Muni Sutta in Sn. had taken place at least a good many years before 
the final edition of Milp. Thus, all these isolated references to Muni- 
gatha and quotations from the Muni Sutta strongly support the internal 
evidence from all sources to establish that the sutta is of great antiquity. 


36. Mrs. Rhys Davids, Outlines of Buddhism, p. 103. 

Pali Buddhist Review 2, 2 (1977) 



N. A. Jayawickrama 
Five Sutta^j of Popular Character 


Parabhava Sutta 

The Parabhava Sutta and the four other suttas which are discussed here 
belong to a stratum of popular Buddhism, and they emphasise the 
practical side of Buddhism, laying down secular advice. The Parabhava 
Sutta is presented as a dialogue between a deity and the Buddha wherein 
para+ -\/bhu the causes for men 's downfall are enumerated by the Buddha 
(pard+\/bhu: defeat). Though there is no deep philosophy underlying 
this sutta its advice is based on high ethical principles. The vices and 
evils denounced by Buddhist and contemporary Indian society are port- 
rayed here as in the Vasala Sutta. It not only reflects the attitude of the 
age towards social evils such as the lack of filial piety, disrespect for elders 
and virtuous men,miserliness, arrogance, addiction to wine, women, and 
gambling and general unchastity, but also serves as an index to what 
was considered wrong in man's dealings with other men right down the 
ages in Indian society. These very sentiments are expressed and repeated 
over and over in numerous other works of Indian literature, especially the 
Dharmasdstras and Dharmasutras, and the sutta is characteristically Indian 
but not merely Buddhistic. The highly ethical basis underlying the sutta 
runs through the whole poem. The Mahdmangala Sutta which lays down 
in the form of "Blessings" the good qualities one should practise is more 
Buddhistic in its values than this sutta, though the two poems taken 
together are complementary to each other as they are based on the same . 
ethical principles. The fact that this poem was meant for the common 
man is seen clearly from the last pdda of the concluding stanza, which 
speaks of a sivam lokam as opposed to sivam padam, the synonym for 
Nibbana. The word ariya (Sn. 1150) has a wider application than the 
normal Buddhist term. 

The language of the sutta is generally archaic. The noteworthy pecu- 
liarities are: — the historical infinitive putthum (Sn. 91c), the historical ppr. 
gen. sg. parabhavato in the refrain, the adjectival form-vf/awo (Sn. 92ab), 
the word bhavam ("worthy" — Sn. 92a) used as in (Skr. bhavdn), the 
primary adjective dessl(\/dvis — Sn. 92d), the verb roceti(Sn. 94c) formed 
after verbs of Class X, the agent noun anutthatd (Sn. 96b), the dialectical 


form pahu in the phrase pahu santo (being able or capable of pra+ s/bhu, 
Sn. 98c, cp.pahutaSn. 102a, etc. and in frequent use in the Canon, specially 
in cpds.), the shorter form sam-in the cpd. sannatim (Sn. 104c, cp. Skr. 
svam— besides svaka, P. saka, also cp. schi — Sn. 108a; sam and sehi are 
poetical forms rather than dialectical variations), the contracted dialectical 
formposo (Sn. 1 10a; vide Geiger§ 30.3), the contracted form issa (Sn. 1 10c), 
the verb supati (Sn. 110c, cp. supina—Sn. 360, etc.) and the uncontracted 
verb of Class X, patthayati (Sn. 1 14c). All these forms show that the 
language of the sutta is rather old. It is also evident that there is an 
abundance of pure -poetical forms as distinct from the normal canonical 
idiom and that the diction of the whole sutta is highly poetic. The poetical 
forms of interest—dhammadessi (Sn. 92d), kodhapahndno (Sn. 96c), 
timbarutthanim (Sn. 110b), etc. The verb interposed between the sub- 
stantive and the adjective , e.g. lokam bhajate sivam (Sn. 115d), khattiye 
jdyate kule (Sn. 114b) etc., the disjunctive employed between the sub- 
stantive and the adjective t.g.pwisam vdpi tadisam (Sn. 1 12b) or even the 
position of the demonstrative adjective in the refrain of the stanzas attri- 
buted to the Bhagava, i.e. pathamo so pardbhavo, etc. are all characteris- 
tic of the poetic language. 

The style of the sutta is neither heavy nor ornate. Though the stanzas 
are highly antithetical, their style is swift and vigorous. Poetical devices 
such as simile, metaphor or pun are few, and in fact there is only one 
metaphor in the whole poem: i.e. Sn. 110b. No definite inference can 
be drawn from the metre of this poem. The 25 stanzas are in Anustubh 
Sloka. There is anacrusis in Sn. 91c and even quarters of the Vipuld-type 
are found at Sn. 91a, 102a, HOac, 112a and 114c. 1 The vigorous Sloka 
metre is best adapted to narrative or dialogue ballads. Doctrinal Develop- 
ments here are almost negligible, but the word anutthatd reminds one of 
the positive concept utthdnaviriya, a term of early doctrinal import. All 
the available evidence from language, style and metre suggests an early 
date for the sutta. The archaic language rich in historical forms, both 
verbal and nominal and containing dialectical variations, the free and 
easy style and the old poetic diction unmarred by any artificial poetic 
devices are in full accord with its early origin. 

External Evidence may yield some data regarding a relative date. The 
comprehensive code of Moral Law promulgated by Asoka has a great 
deal in common with the Parabhava, Vasala and Mangala Suttas. Al- 
though Mookerji 2 is emphatic that Asoka 's ' 'Dharma" was not Buddhism 
but his own ethical philosophy, the strange s imilarity of idea s in his code 

1. Helmer Smith, Sn.A. 640-641. 

2. Radhakumud Mookerji, Asoka, p. 68, Gackwad Lectures, 1928. 

88 Pali Buddhist Review 2, 2 {1977) 

of ethics and in these sulfas is conclusive evidence of the connection 
between the two. Seeing how far he is influenced by Buddhism and 
Buddhist literature the inference that he based his code on sulfas such as 
these are similar literature is quite justifiable. The fact that Asoka not 
only is acquainted with the ideas here, but also inculcates them is proof 
of the popularity of secular ethics of this type. 


Vasala Sutta 

The position of the Vasala Sutta in the vagga immediately after the Para- 
bhava Sutta leads one to the natural inference that an attempt has been 
made at an arrangement of the suttas according to their subject-matter. 
Such instances are quite frequent in the Canon. The two suttas, Parabhava 
and Vasala agree with each other in subject-matter, style, language and 
metre while the two differ in details regarding the outward form. The 
Vasala Sutta falls into the "Akhydna "-class, though the Parabhava Sutta 
cannot be strictly called so. The gat has of the former can form an in- 
dependent sutta without the brahmin being introduced to it at all, but the 
latter is a pure dialogue like the Kasibhdradvdja Sutta. In contents the 
two suttas agree very closely. Lack of filial affections is deplored in 
identical words (Sn. 98, 124) and so is deception practised on brahmins 
and holy men (Sn. 100, 129). The four major evils of killing, stealing, 
falsehood and adultery are condemned in Sn. 117-123 (Vasala). False- 
hood is referred to in Sn. 100 and adultery in Sn. 108 (Parabhava). Both 
poems deal with anger (Sn. 96, 116, 133), pride and arrogance (Sn. 104, 
132), miserliness and lack of hospitality (Sn. 102, 128, 130) and various 
other social evils. The same subject is dealt with in identical words in 
two instances (quoted above— Sn. 98, 124; 100, 129). The Vasala Sutta 
deals more fully and in a more comprehensive manner with most of the 
subjects taken up in the Parabhava Sutta; and mentions more vices and 
evil practices than the latter. Though both suttas are true to the spirit 
of early Indian ethics, the Vasala Sutta goes a step further in emphasising 
that one's own actions alone qualify one for condemnation and not one's 
birth (jati). The gathd, 

Najacca vasalo hoti, najacca hoti brahmano 
kammand vasalo hoti, kammand hoti brahmano. 

(Nothy birth does one become an outcaste or a brahmin, but by one's 
action one becomes an outcaste or a brahmin) occurs twice in the sutta, 
and an illustrative anecdote is appended. The gathd clearly conveys the 
Buddhist attitude to caste and the note struck here is truly Buddhistic. 

Pali Buddhist Review 2, 2 (1977) 89 

The sutta itself can be divided into four parts:- 

I The prose introduction the prelude to the sutta. 
II. Thebodyofthe««fl(S«. 116-136) containing the aphoristic sayings 
dealing with the various vices and evil practices. 

III. The short dissertation on kamma (action) including the colourful 

illustration of Matanga. 

IV. Aggikabharadvaja's confession of faith in prose, forming the 

conclusion of the sutta. 


The sutta can retain its characteristics and form a coherent whole with- 
out parts, I, III and IV and yet be called Vasala Sutta. Sn. 136 appears 
as a crescendo and concluding verse of the sutta. This is further strength- 
ened by Sn. 135 which, in addition to its extra pddas sums up the categories 
of vasalas in its last line, 

Ete kho vasala vutta, maya vo ye pakasitd 

(These whom I have declared unto you are vasalas). This summing up 
may be compared with Sn. 269, the concluding gathd of the Mahdmangala 
Sutta In both instances the refrain occurs up to the gathd immediately 
preceding the respective stanzas, and thus Sn. 135 provides a suitable 
conclusion to the sutta. The next stanza too, which in a dramatic manner 
breaks down the age-old barrier of caste and attributes baseness to base 
actions rather than to birth, probably belonged to the original sutta. The 
illustration (nidassana) that follows appears as a separate sutta or as a 
separate section appended to the sutta at a subsequent date. The position 
of these six stanzas at the end of the sutta makes this suggestion very 
plausible. The repetition of Sn. 1 36 at Sn. 142 is merely for the purpose 
of emphasising this essentially Buddhistic aphorism. It also provides 
a suitable climax to the enhanced sutta. 

There is no doubt that the episode of Matanga is borrowed from popular 
tradition. The story Matanga occurs in the Anusdsanika-parvan of the 
Mahdbhdrata (Mbh. XIII, 3, 198 ff), but it differs considerably in details 
fromthatinSn. Both Sn. and Mbh. agree on his lowly birth (C^layon- 
yam jdto, Mbh, XIII, 3, 198). The outline of the legend in Mbh. is:- 
' 'Matanga, son of a brahmani was informed by a she-ass that he was in 
reality a canddla, and in vain tried by way of penance to become a brahmin, 
at last he 'succeeded in becoming Candodeva" * The ex sitence o a 
parallel legend in Mbh. need not necessarily imply that either was ba ed 
on the other. The probability is that both versions go back to anearlier 


s.v. Sorensen, Index to the Names in Mahabharata. 

90 Pali Buddhist Review 2, 2 (J 977) 

tradition (probably oral) and the two as they are, represent parallel 
developments. (Also cp. Matanga Jdtaka). 

On the other hand, the position of Sn. 124 and 129 seems different. The 
topic under discussion in Sn. 125 is cruelty by word or deed to one's own 
kith and kin. It seems probable that the connected idea of not supporting 
one's aged parents has been transported here, and the stanza borrowed 
wholesale. Similarly, Sn. 129 appears as an interpolation. The stanza 
that immediately precedes it (Sn. 128) denounces the action of the person 
who does not return hospitality to his erstwhile host; and the stanza that 
follows it (Sn. 130) condemns the person who, instead of feeding them 
abuses brahmins or samanas who come to his door at meal-time. Both 
these stanzas deal with the feeding of guests or mendicants, but Sn. 129 
speaks of the deception practised on mendicants, religious or otherwise, 
by uttering falsehood. Although Sn. 129 disturbs the logical trend of the 
two stanzas on either side of it, the occurrence of the phrase, yo brahmanam 
va samanam va in Sn. 130 seems to have been considered sufficient reason 
to introduce Sn. 129 which incidentally begins with the same phrase. The 
inference that Sn. 124 and 129 are interpolations implies that the Parab- 
hava Sutta is earlier than the Vasala Sutta. This need not necessarily 
be so. The only legitimate conclusion is that the final redaction of the 
latter took place after the composition of the former. The position of 
these two stanzas in the two suttas sheds some light on this point. In the 
Parabhava Sutta, these two stanzas dealing with similar topics, occur as 
consecutive answers given by the Bhagava, whereas in the Vasala Sutta 
they are separated by four other stanzas, two of which (i.e. Sn. 126, 127) 
deal with a different topic altogether. 

The similarity of ideas in the two poems does not call for particular 
attention on account of the fact that they deal with parctically the same 
subject, Language, style, metre and syntax too do not help in determin- 
ing the age of the two poems in relation to each other. It is solely on the 
data provided by these two stanzas and the occurrence of the illustrative 
episode of Matanga, (when the sutta proper could end at Sn. 135ef where 
the categories of vasalas are summed up, or at Sn. 136 which provides a 
fitting climax) it can be said that the sutta may have undergone a change 
at the hands of a subsequent editor. The stanzas Sn. 137-142 appear as 
a subsequent addition made by a later editor. It is quite probable that 
the earliest form of this sutta did not include these six stanzas, Sn. 124 
and 129, nor perhaps the prose sections. Judging from internal and 
external evidence the earliest versions of both suttas appear contemporary. 

As stated earlier the language of the gathas is quite similar to that of the 
Parabhava Sutta. There are old historical forms like the opt. 3 sg. jahna 

Pali Buddhist Review 2, 2 (1977) 91 

(Sn. 116d-134d), the denominative mamayitam (Sn. 119b), shorter inst. sg. 
theyya (Sn. 119c — Vedic steya nt.), the ppr. pass, cujjamano (Sn. 120b), 
the contracted verb adeti (Sn. 121c) besides ddiyati (Sn. 1 1 9c) the absolutive 
bhutvana (Sn. 128b — also poetical), the ppr. medial of the desiderative 
nijigimsano (Sn. 131c), the pronom. adj. inst, sg. sena (Skr. svena, cp. sam 
Sn. 104c, sehi Sn. 108a, 132c), the inst. sgg. jaccd (Sn. 136ab, 142ab), 
duggaccd (Sn. 141d), kammana (Sn. 136cd, 142cd) and amina (Sn. 137a cp. 
also amuna) and 3 pi. A. pada upadissare (Sn. 140d). Besides the poetical 
forms like bhutvana, upadissare, sena, etc. there occur in this sutta as in 
the previous one many cpds. e.g. pdpamakkhi (Sn. 116b), vipannaditthi 
(Sn. 116c), paticchannakammanto (Sn. 127c), etc. The sutta preserves the 
old Pali idiom, e.g. pane daya, (Sn. 117c), yam paresam mamayitam (Sn. 
119b), etc. Often the same idiom, is seen to occur in canonical prose, 
e.g. attahetu, parahetu, dhanahetu. .(Sn. 122ab), akincikkhakamyata (Sn. 
121a, cp. labhakamyata). There is an irregular ace. sg. of the ppr. vajatam 
(Sn. 121b, v. 1. vajantam vide Geiger, § 130; the Corny, explains it as gacch- 
ctntam — SnA. 179). There is also an abundance of Vedic enclitics like 
ve and ha-ve (vai and ha vai). All these characteristics of old Pali and the 
general diction of the poem which is archaic suggest an early for the sutta. 

The Style and metre of the poem are similar to those of the Parabhava 
Sutta. The metrical irregularities are few; i.e. odd quarters at Sn. 118a, 
121a and 123c and an even pada at Sn. 124c. 4 Evidence from language, 
style and metre shows that the two poems are contemporary, though on 
careful examination some parts of the Vasala Sutta appear to be younger 
than the Parabhava Sutta. 

No doctrinal developments as such are noticeable in the sutta. How- 
ever in spirit it is more Buddhistic than the former. The four major evils 
of killing, stealing adultery and falsehood have already been noticed to 
occur in Sn. 117-123, in same order as the first four precepts, in addition 
to the other allied misdeeds as highway robbery and plunder. In spite 
of the popular nature of the sutta the occurrence of the two terms, dittheva 
dhamme and sctmparaye (Sn. 141ab) suggests some development in Budd- 
histic ideas; but these terms are of no great value as they are equally 
common in early Indian literature. The words vipannaditthi (Sn. 116c), 
moha (Sn. 131b), araha and anaraha (Sn. 135ab) are not used in their 
specialized meaning as found in Buddhism. It is significant that Sn. 
134b speaks of the savakas and not of the Sangha, and it is probable that 
this sutta is quite distinct from monastic Buddhism. The phrase khattiya 
brakmana in Sn. 138, like the canonical phrase samana-brahmana, un- 
consciously suggests the order of precedence as the Buddhists conceived 

4. Helmer Smith, ibid. 


Pali Buddhist Review 2, 2 (1977) 

it, 5 though the sutta itself repudiates the stigma attached to low birth. 
The mention of devayana, the path leading to the devas in Sn. 1 39 shows to 
some extent how far the sutta echoes the then-current Indian thought. It 
is clear that the goal aimed at is brahmaloka. The earlier Upanisads 6 
mention the two paths by which a departed soul proceeds to enjoy the 
fruits of his actions. They are the devayana or the arcirmarga the path of 
light leading to the plane of Brahma or satyaloka and the pitryana or the 
dhumamarga, the path of darkness leading to the region of the moon or 
candraloka. When Indian thought evolved and gradually established the 
identity of Self with Brahman, devayana became the path leading to the 
union with the Highest. 7 It is not clear what stage of development in 
Indian thought Sn. 139 reflects, yet the final goal mentioned is brahma- 
loka. Peshaps sivamlokam (the world of happiness at Sn. 115) also refers 
to the same state. 8 The Commentator rightly interprets it as devalokam 
(Sn.A. 173). Both these reference show that these poems are not doctrinal 
dissertations but suttas meant for the inculcation of popular ethics. 

All the available external evidence too shows that the sutta belongs to 
the realm of popular ethics. It contains ideas common with the Epics 
and other Sanskrit literature. Sn. 122 may be compared with Manu. VIII, 
13; Sn. 128 with Mbh. XIII, 126; 27; and Taittiriyasiksavallll\,2; and Sn. 
135 with Manu. X, 12; 16 and 26, etc. The observations made with 
reference to the Parabhava Sutta that Asoka's moral code apply here too. 
The gathas appear to be very old in the light of internal evidence and the 
testimony from the Edicts suggests that they should be at least pre-Asokan. 


Mah? mangala Sutta 

It has already been stated that the Mahamahgala Sutta (also known as 
the Mangala Sutta) is complementary to the Parabhava Sutta. Both suttas 
contain a short prose introduction with identical words, and a devata is 
introduced as the Buddha's interlocutor. The only difference in form, 
between the two suttas is that the Mangala Sutta unlike the other, is not a 
proper dialogue, for, the devata is represented as asking only one question 
to which the Buddha gives an uninterrupted reply. The two suttas 
categorically state the various factors which lead to one's downfall and 
which are considered as blessings respectively, and conclude didactically 
summing up the enumeration. A regular feature in the poems is the 

I' ^L S P UmC l Kh ?f' iy ^' Br f h ™* a > Ves ™> Sudda occurs many times in M. and D 

6. Brhadaranyaka Upamsad, VI. 2, 2; IV. 11, etc. u 

7. Also vide Radhakrishnan, Indian Philosophy, Vol. I, pp 252-255 

8. As stated, this term cannot refer to nibbana as the mention of a ' 'nibbana- 
oka anywhere in the Canon. It is either nibbana pada or nibbana dhatu ° ma 

Pali Buddhist Review 2, 2 (1977) 


refrain which is a feature in the greater part of the Vasala Sutta. The sutta 
lays emphasis on good living and gives practical advice. It is essentially 
didactic like the gnomic poetry of the Sanskrit literature. The word 
mangala conveyed to the brahmins anything that was considered as auspi- 
cious. In every society, whether past or present, Occidental or Oriental, 
there are certain beliefs and superstitions to which people attach some 
importance in varying degrees. Mangala to a brahmanical society in 
ancient India represented all the sights and sounds, actions, ritual and 
ceremonies which they deemed holy or auspicious. In this sutta the 
Buddha is seen giving a new value to the term mangala employing it to 
stress the importance of a righteous living. The sutta does not attempt 
to teach anything new, but inculcates in a different form the ethical 
principles already known to the Indians. There is no deep philosophy 
underlying the sutta, yet it has to some extent a Buddhist background. 
The theme, kalena dhamma savanam (Sn. 265c, listening to the dhamma 
at the proper time) kalena dhamma-sakaccha (Sn. 266c, religious discussions 
at the proper time), samananam ca dassanam (Sn. 266b, paying homage 
to the monks), ariyasaccana dassanam (Sn. 267b, an insight into the ariyan- 
noble truths) and nibbanasacchikiriya (Sn. 267b, the realization of nibbana) 
are decidedly of Buddhistic application though dhammacariya (Sn. 263a, 
living in accordance with the dhamma), appamado ca dhammesu (Sn. 264c, 
perseverance in doing good deeds) and patirupadesavasa (Sn. 200a, living 
in a suitable region) are capable of being given a wider interpretation than 
suggested by the Commentator (Pj. I. 123-157). The perfect balance of 
mind under all conflicting circumstances. (Sn. 268) is again a characteristi- 
cally Buddhist concept. The sutta thus is essentially Buddhistic although 
it deals with popular ethics. 

This sutta occurs verbatim in the Khuddaka^patha and the Paritta-patha? 
It is also one of the Tun-sutraya (the Three Suttas, the other two being 
Ratana and Metta Suttas), used at Pirit ceremonies; which shows that the 
sutta has enjoyed great popularity from comparatively early times (when 
the Khp. was compiled); up to the present day. 10 There is a.jataka known 

9. The Paritta-patha is a collection of suttas varying in number from 28 to 32, 
taken from various parts of the Canon. It is known in Ceylon as the Pirit-pota (the 
Book of Pirit). Also vide Winternitz, History of Indian Literature, Vol. II, p. 80. 

10. Khp. as a collection cannot be of very early date. From the negative evidence 
that no mention of it is made in the Canon or in Miln. it may be inferred that it came 
into being sometimes later, though argument from silence is not always very satisfactory 
evidence. On the other hand, Miln. mentions Sn. by name (vide U.C.R.. Vol. VII, 3), 
and it is quite probable that the author of Miln. know Sn. as it exists to-day. Miln. 349 
mentions Khuddaka-bhanaka, but this is no evidence for the existence of Khp. It only 
refers to a collection of minor pieces, probably the greater part of what is now known 
as Khuddaka Nikaya, just in the same way as Digha or Majjhima-bharaka referred to 
the reciters of long or medium sized suttas which were invariably included in the Digha 
and Majjhima Nikayas respectively. Thus, the earliest collection in which Mangala 
Sutta was included is probably Sn. though it may have existed earlier as an individual 
sutta. , 

94 Pali Buddhist Review 2, 2 (1977) 

as the Mahdmangala Jdtaka (No. 453) which the Commentator (J. IV, 
72-73) associates with the Mangala Sutta and quotes the opening pada 
of the devatd's question mentioning that the total number of mahgalas 
enumerated in the sutta is 38; but it has no connection whatsoever with 
the present sutta, and it is in reality a dissertation on happiness in accord- 
ance with Brahmanical ideas of life rather than Buddhistic principles. 

There are no specific linguistic forms in the sutta that may be classed as 
very ancient, nor are there signs of lateness in the language. It is the 
normal Gatha-Pali idiom with the usual poetic diction. The stanzas 
are highly rhythmical and melodious. There is no involved syntax and 
the language is simple. The few linguistic forms which call for attention 
are: — acintayum (Sn. 258b) the historical Aorist 3 pi., sotthanam (Sn. 258c) 
ace. sg. of sotthanctnt. cp. Skr. svastyayana, sovacassatd (Sn. 266a) abstract 
of the secondary form from su-vacas and the usage of dassanam (Sn. 266b, 
267b) in its literal and applied meanings of visiting to pay homage and 
unsight into (vision of), respectively. The phrases matapitu-upatthanam 
(Sn. 262a), ariyasaccdna-dassanam (Sn. 267b) and sabbattha-m-apardjitd 
(Sn. 269b) betray the flexibility of sandhi in Pali, specially in metrical 
exigencies. The metre of the poem is Anustubh Sloka, and the few metrical 
irregularities are: one instance of anacrusis at Sn. 260a and two instances 
of even quarters at Sn. 260c and 265c. The sutta contains a few special 
Buddhistic terms in addition to those that are in common with contempor- 
ary Indian religious systems. Ariyasaccdna-dassanam (Sn. 267b) is a 
definite reference to the Noble Truths of Buddhism, and nibbdnasacchi- 
kiriyd (Sn. 267c) is the attainment consequent on the obtaining of an 
insight into the Ariyan Truths. Other concepts such as tapo (ascetic 
practices) brahmacariyd (celibacy) Sn. 267a, attasammapanidhi (& thorough 
development of personality — Sn. 260c), khanti (forbearance — Sn. 266a) 
etc. are of general Indian origin and therefore are of no special importance. 

External Evidence consists mainly of a comparison with the Moral Law 
of Asoka promulgated in the Edicts. Asoka's dhartna, like the sayings 
in the three suttas, Vasala, Pardbhava and Mangala, is not any religious 
system peculiar to one sect or school, but contains practical and doctrinal 
advice embracing the various relations of life. However, a close com- 
parison shows that Asoka had drawn his material from a literature very 
similar to these suttas. From his acquaintance with certain parts of the 
Canon i.e. the seven Dharmaparydyas some of which have been traced 
to Sn. (vide PBR, 1, 3, p. 137) it may be inferred that he was equally 
acquainted with these suttas. The following table 11 shows to what 
extent the contents of these suttas can be compared with Asoka's dharma: 

11. This table is based on Radhakumud Mookerji's analysis of the Moral Law of 
Asoka in his Gaekwad Lectures, pp. 69 ff. 

Sutta Nipdta: 
Sn. 259b, panditdnam ca 

Sn. 259c, pujd ca pujaney- 

Pdli Buddhist Review 2, 2 (1977) 95 


1 . Obedience to (a) elders R.E. IV, 12 (anu- 
pratl patipati) P.E. VII, (b) teachers R.E 

2. Respect (a) of pupih-antevasi-tow&rds 
their gurus M.R.E. II, (b) towards gurus yanam 
R.E. IX. 265a > garavo. 

3. Proper treatment towards (a) ascetics, Sn. 100, 129, 130. 

both brdhmana and samana R.E. IV, P.E. Sn. 98-124, 102, 104c, 125. 
VH, (J) relations M.R.E. II, R.E. IV, XIII. 

4. Charity (ddnam) R.E. Ill, VII, VIII, IX, Sn. 263a, 102c. 

5. Abstention from slaughter of and Sn. 117-118. 
violence towards living beings, R.E. 1TI, 

IV, XI, P.E. VII, R.E. IX, IV, P.E. VI1LR.E. 
XIII, M.R.E. II, cp. R.E. Ill, IV, IX, XI, 
XIII and P.E. VII. 

6. Kindness (day a) P.E. 11,-VII. Cp. Metta Sutta, Sn. 143- 


7. Truthfulness (satyam) M.R.E. II, P.E. II, Sn. 122c, 100c, 129c. 

8. Gentleness (mdrdavam) R.E. XITT, P.E. 


9. Gratitude (krtajhatd) R.E. VII. 

10. Attachment to dharma (Asokan mora- 
lity) R.E. XIII. 

11 Purity of heart (bhdva suddhi) R.E. VII 

Sn. 143d. 

Sn. 265b. 

Sn. 92c, 263a, 264c, 265c, 

Sn. 260c, cp. yakkhassa 
suddhi Sn. 478, 876. 

Of the requisite qualities mentioned in P.E. I for the attainment of happi- 
ness in this world and the next, dharma-kdmatd occupies the first place 
(No. 10 in Table). Susrusi (obedience), No . 1 in table bhaya (fear to do 
wrong— cp. ottappa in Pali, Sn. 133c, etc.), and utsdha (effort— cp. utthdna- 
viriya, see anutthdtd, Sn. 96b) are three others. In R.E. XIII Asoka 
summarises his 'Dharma' as 1. Aksati (non-injury— cp. Sn. 117-118) 2. 
samyama (restraint quite frequent in the Mum-Ballads of Sn. cp. Sn. 264, 
etc.) 3. samacaranam (impartiality) and mdrdavam (gentleness; No . 8 m 
table). On these and numerous other points (enumerated by Mookerji, 
Asoka, pp. 69-78) Asoka's "Dharma" bears a strongj-e semblance t o the 
Rock Edict, P.E.: Pillar EdictTM.R.ETTMinor Rock Edict. 


Pali Buddhist Review 2, 2 (1977) 

ideology of these suttas. Although one may not be quite certain of the 
existence of Sn. as a collection in Asoka's time, there is no doubt that 
many of the suttas constituting Sn. were anterior to Asoka. It has already 
been noticed that Asoka had been influenced by a literature quite similar 
to these popular ethics. The internal evidence reveals that the suttas 
are old, andthat they preserve definite characteristics of the poetic language 
which probably preceded the standardised canonical Pali prose. It is 
thus highly probable that these suttas were known to Asoka. 


Metta Sutta 

The Metta Sutta, another member of the triad of suttas, is 1 also found in 
Khp. as No. 9. The Commentator's introduction which is historically 
of a lower level of interpretation states that the sutta was preached by the 
Buddha as a topic of meditation to monks and to serve the purpose of a 
paritta to ward off clangers arising from evil spirits (Pj. I. 231-232). There 
is no doubt that the sutta provided a useful topic of meditation for both 
monks and laymen, and in subsequent years, even as early as the com- 
mentarial epoch, or perhaps earlier, it was used as a paritta. Its inclusion 
in Khp., a handbook of popular Buddhism, indicates that this sutta, like 
its two companion suttas, was very popular from comparatively early 
times. The sutta inculcates the practice of metta (amity, or love to all 
beings). The theme is an early tenet of Buddhism and the idea corresponds 
to dayd of the Sanskrit epics and other allied literature. Even before the 
four Brahma-viharas were fixed and standardised the term metta is to be 
met with in association with such concepts as upekha, karuna, vimutti 
and mudita (Sn. 73). The theme is developed from various aspects in the 
sutta: — (a) Wishing happiness and well-being to all creatures irrespective 
of their size or form or stage of growth (bhuta va sambhavesi va: creatures 
come into being or in their embryonic state, Sn. 147c). (b) The negative 
aspect of the absence of ill-will towards them (Sn. 148). (c) The develop- 
ment of boundless thoughts of love, as deep as maternal affection to all 
creatures (Sn. 149). (d) The diffusion of unobstructed thoughts of loving 
kindness in all directions at all times (Sn. 149-150). It is extolled as the 
brahma-vihara — the highest abiding 13 (Sn. 151). The 10 opening lines of 
the sutta are of an introductory nature. They describe the atthakusala — 
he who is bent on his welfare. Though santam padam — tranquilled state 
(Sn. 143) — need not necessarily always signify nibbana, the qualifying 

13. Mrs. Rhys Davids (in "What was the Original Gospel of Buddhism?") thinks 
that this line preserves "a metrical legacy" of the disciples of an unknown Brahmin 
teacher, and sees God in the epithet brahma. She translates Sn. 151d as, 
"God have they here this living called". 

Pali Buddhist Review 2, 2 (1977) 97 

remark that the atthakusala should be detached from family life (Sn. 144d) 
suggests that the poem was primarily meant for the monks. A descriptive 
classification of creatures is made at Sn. 146 and another mutually ex- 
haustive and more precise division at Sn. 147c 14 . The development of a 
mental attitude (manasam bhavayati, cp. Sn. 149d, 150b) consisting of the 
thoughts of love is the keynote of the poem and it concludes with an ex- 
hortation to make an end of birth (Sn. 152d). 

As far as linguistic evidence goes what has been said of the other suttas 
applies here too. There is ellipsis in the opening stanza of the sutta (Sn. 
143). The form addifthd at Sn. 147a has -dd- either metri causa or as a 
consonantal doubling after the negative prefix; cp. appasada. The 
indefinite adverb katthacmam (Sn. 148b) contains a contamination of two 
indefinite suffixes cid and cana. There are two forms with the contraction 
of the final -aya>-d viz. vydrosana and patighasanna: (Sn. 148c). The 
regular form niya for Vedic nija (also P. nija) through Prakrit nia (vide 
Geiger § 36) occurs. Besides these there are other forms sayano (Sn. 151b) 
historical ppr. medial, the affirmative particle jdtu (Sn. 152d) which is 
practically confined to poetry, Vedic forms as dyusa (Sn. 149b) and a 
wealth of optative forms illustrative of various types used in Pali; viz. 
3 sg. assa (Sn. 143d), 3 sg. samdcare (Sn. 145a), anurakkhe (Sn. 149b), 
bhavaye (Sn. 149d, 150b), medial 3 sg. nikubbetha (Sn. 148a) kcheyya (Sn. 
148d). adittheyya (Sn. 151c) and 3 pi. upaddaveyyum (Sn. 145b). There 
also occurs an inorganic sandhi at Sn. 151d. viz. idha-m-ahu. 

The style of the sutta is free and easy to a great extent and the ideas are 
expressed lucidly. The sutta being didactic, the greater part of it is 
explanatory and injunctive. Though it contains two long lists (Sn. 143c- 
144d, 1461-147d) no laboured effect is produced, as there is an easy flow 
of words along with its rhythmic effect. There is only one simile in the 
whole poem (Sn. 149ab) and it appears quite apt, as it emphasises the 
central theme. The poem is written in a metre described as Aryd (Giti) 
by Helmer Smith. 15 Yet the metre here is not the proper classical Giti 
or any of its sub-types including Aryd. Normally the syllabic instants of 
the first and third pddas of the Giti metres are limited to 12, whereas the 
other two pddas vary from 15 to 18. None of these 10 stanzas corresponds 
to any of the varieties of the classical Giti metre, and at best what is found 
here is a very free modification of the Arya-Giti. the syllabic instants of 
the Metta Sutta vary from 11, 17, 13, 17 in Sn. 152 to 16, 17, 15,J8_JnSK. 

14. It is significant that the classification into the five types jalabuja, andaja, sam- 
sedaja and opapatika (viviparous, oviparous, moisture born and of spontaneous Dinn; 
a division known to be in use comparatively early in India, does not occur here, it may 
perhaps indicate that the Buddhist writers had not yet adapted it in their works Curing 
the time of these suttas, though the classification occurs in prose sections ot the <_anon 
(D. III. 2: 0, M. I. 73 S. Ill, 240, etc.). 

15. Op. cit. (Sn.A.. 637). 


Pali Buddhist Review 2, 2 {1977) 

147. In two instances the number of matras (syllabic instants) in a full 
stanza is less than 60, in two 60 (the prescribed number in Sanskrit poetics) 
and in the other six over 60. However it is worth noting that this metre 
stands "between the more archaic technique of the sloka-tristubh and the 
jagatis, atijagatis, etc, of the late-canonical and semi-canonical com- 
positions". 16 

Doctrinal developments. In. its tone and outlook the sutta bears signs 
of antiquity. The term atthakusala (Sn. 143a) probably refers to the 
attha (weal) pertaining to both this world and the next. The emphasis on 
a life which is beyond the censure of the wise (Sn. 145ab) may suggest that 
the attha was primarily connected with this world. The Commentator, 
however, interprets santam padam (Sn. 143b) as nibbana (Pj. I. 236) merely 
because the term is used as a synonym for the latter. Yet, the state of 
mental tranquillity referred to here is capable of a more general inter- 
pretation. This is further supported by Sn. 143cd-145ab, which do not 
apply to one who has attained nibbana: and therefore attha should be 
interpreted as benefit here on earth rather than well-being after death or 
even paramattha the summum bonum. 

The ideal envisaged in Sn. 144 is that of the ascetic or the muni. This 
certainly reflects early material. The next two lines which allude to 
public opinion are not wholly inconsistent with the Muni-idea.\, though 
as a rule the muni is not influenced by the outside world (cp. Sn. 213b, 
214b, etc.). This difference in attitude does not imply a later phase of 
development, nor does it indicate any real departure from the ideal in 
early Buddhism. The probability is that the sutta was meant for a wider 
circle than ascetics alone, and the authority of the vinhu had to be reckoned 
with in preparing one's self for the tranquil-state of mind which would 
be the basis for the contemplation (manasam bhaveti) on metta. The 
concept of metta is suggested to be of pre-Buddhistic origin, 17 but Buddhism 
and its senior contemporary religion Jain ism were responsible for the 
widespread movement of not only non-injury to living things, but the 
actual practice of amity towards them. The phrase manasam bhaveti 
(develop a frame of mind; cp. manisa Rv. X. 129) is not in frequent use 
in the language, and the word manasa used absolutely is semi-technical 
in character (s.v., PTS). Neither the concept metta nor the term manasa 
yields any conclusive data. The two words ditthi and dassana (Sn. 152) 
are used as mutually contradictory terms from the earliest times: 18 they 
are diametrically opposed to each other in their semantic development 

16. Ibid. Helmer Smith. ' 

17. Mrs. Rhys Davids, op. cit. 

1 8. Mrs. Rhys Davids in her translation of Khp. in the Minor Anthologies considers 
this stanza late for reasons she adduces in her introduction. 

Pali Buddhist Review 2, 2 (1977) 99 

though they are derivatives from the same root (drs). There is hardly an 
instance ofditfhi being qualified zssamma or miccha in all the 45 occurren- 
ces of the word in Sn. The term ditthi embraces all philosophical views 
and speculations which were (perhaps later) designated as micchaditthi, 
while dassana is a more precise concept signifying insight. The latter 
has no bearing on the question of the difference between sammaditthi and 

External evidence— The only form of external evidence available is 
the occurrence of similar or parallel ideas in other works. The idea metta. 
occurs frequently in the Canon in the list ofBrahmaviharas, and also singly 
in various other contexts, e.g. metta ceto-vimutti at D. 1. 251, $.11. 265, 
A. IV. 150, It. 20, etc., metta-sahagatena cetasa at D.l. 250, etc., S.V. 115, 
A. 1. 138, etc., mettam cittam at D. I. 167, 111. 237, Sn. 507, Vin. II. 1 10, 
A. II. 72, Th\. Th2, etc. and metta-bhavana at Miln. 199. At S. I. 75 
Pasenadi Kosala declares to Maliika that the dearest thing to a person 
is his own self. At the end of the conversation they go to the Buddha who 
advises them 

Sabba disanuparigamma cetasa 
rc'ev' ajjhagama piyataram attana kvaci, 
evam piyo puthu atta paresam 
tasma na himse par am attakamo. 

(Having mentally surveyed all directions 1 have not found anywhere, any- 
thing so dear to me as my own self. So is it to Jhe others that each one's 
self is dear. Therefore let him who loves his own self not bring harm 
upon another). Cp. Sn. 705, Dh. 129, 1 30. Here the standard of judgment 
in refraining from injury to others is one's love for one's own self. The 
same idea is reflected in Yajnavalkya 's advice to Maitreyi in Brhadaranyaka 
Upanisad (Brh. II. 4: IV. 5) cp. Bhg. VI. 32 which conveys the same idea. 
A passage occurring at Mbh. XI. 7, 1 

Na hy atmanam priyataram kincid bhutesu niscitam: 
anistam sarva-bhutanam maranam ndma Bharata: 
tasma t sarvesu bhutesu day a karya vipascita. 

(Undoubtedly there is nothing so dear to beings as their own selves: indeed, 
death is most unpleasant to all creatures, O son of Bharata. Therefore let 
the wise man extend kindness to all creatures) is quite similar to the passage 
at S.I. 75. 

In all these instances the reason adduced for one to refrain from harming 
others is the love one bears to one's self. On the other hand in the Metta 
Sutta the practice of metta is not prompted by any such motives. It is 
metta for its own sake. It is not possible to ascertain which idea was 

100 Pali Buddhist Review 2, 2 (1977) 

earlier. Though the concept metta occurs frequently in early canonical 
works metta bhavani is scarcely mentioned. As a term metta bhavana 
may be of later growth, yet the idea seems old. The creation of an active 
mental force (manasam) consisting of thoughts of love is fundamentally 
the same as metta bhavana, and is perhaps the predecessor of the latter 
term. 19 

Ratana Sutta 

The popular character of this poem is seen in the opening lines, yanidha 
bhutani etc. The bhutas (spirits) are addressed and their goodwill (sum- 
anas) is invoked. They are requested to extend thoughts of friendliness 
(mettam karotha) to the human race. The naivity and the simplicity of the 
two opening stanzas are reminiscent of the Vedic hymns. This is specially 
true of, 

diva ca ratto ca haranti ye balim 
tasma hi ne rakkhatha appamattd 

(who bring you oblations day and night; therefore protect them arduously). 
Every word of these two stanzas is full of meaning and of echoes. Al- 
though these beings are invoked for protection, the central theme is the 
exaltation of the Three Ratanas, Buddha, Dhamma and Sangha. One 
may see in this sutta a synthesis of popular cults and Buddhism, yet the 
synthesis is very remote, far from being complete. There is no real 
adaptation of Buddhism to popular cults and ritual, but on the contrary 
popular Buddhism has taken for granted a prevalent cult. However, 
with time there sets in a change and this sutta along with many others 
becomes uparitta (a ward-rune) and there-by part and parcel of every-day 
Buddhism. Its inclusion in the "Three Suttas", the Piritpota and the 
Khp. has already been mentioned. 

The invocation of blessings in the sutta is in the form of a saccakriya 
(asseveration by truth) viz. etena saccena suvatthi hotu. A remarkable 
feature of the poem is the evidence of a growth of a complete Buddhist 
doxology. The term Tathagata an epithet often applied to the Buddha, 
is extended to both the Dhamma and Sangha. 20 Seven of the 12 stanzas 

19. Miss I. B. Horner in her Review of Woven Cadences (Hibbert Journal, October 
1945) points out that the formula uddham adho ca tiriyam is found Only at Sn. 150 in 
connection with metta. However, it is found in other passages without any reference 
to metta; e.g. 537, 1055, 1068, etc. 

20. The three formulas used in the Buddhist daily prayer in praise of Buddha, 
Dhamma and Saiigha, viz. hi' pi so bhagava-pe-svakkhato bhagavata dhammo-pe-and 
supatipanno bhagavato savakasangho-pe-respectively , culled from the Canon (e.g. D. II. 
93 fif., III. 5, A. I. 207 ff., 56 ff., IV. 406 ff., etc.) may be compared with this. Also cp. 

Ye ca Buddha (dhamma, sangha) atrtaca, 
ye ca Buddha (dhamma, sangha) anagata, 
paccuppanna ca ye Buddha (dhamma, sangha). 
aharri vandami sabbadd 

Pali Buddhist Review 2, 2 (1977) 


Sn. 224-235 devoted to the Three Ratanas are in praise of the third ' 'Jewel, ' ' 
the Sangha. This may probably indicate a conscious effort on the part 
of the Sangha to assert its importance. The members of the Sangha are 
spoken of as the disciples of the Sugata who are worthy of offerings (te 
dakkhineyya sugatassa savaka — Sn. 227c). They are the recipients of 
dana or yahna in Brahmanic terminology, 21 and as such form an important 
factor for man 's acquisition of merit (anuttaram puhnakkhettam lokassd). 
It is obvious that the sutta reflects a time when there had come into 
existence an organised coenobitic Sangha as opposed to forest dwelling 
anchorites — munayo. The invitation extended to the bhutas to join in 
the worship of the Buddha, Dhamma and Sangha (who are honoured by 
gods and men — devamanussapujita) not only betrays the popular nature 
of the sutta but shows that its composition had taken place when worship 
formed an essential factor in the religion. 

The sutta can be divided into three parts, viz. I. Sn. 222-223, II. Sn. 
224-235, and III. Sn. 236-238. Part I forms the introduction which 
consists of an invitation to the bhutas whose metta is invoked. These are 
the only stanzas of the poem which resemble the Vedic hymns closely. 
As in the Vedic hymns (a) the opening lines form the invocation (Sn. 
222ab), (b) a request is made for their goodwill (Sn. 222cd), (c) their 
metta is solicited (223b"), (d) they are reminded of services rendered to them 
(Sn. 223c) and finally (c) their protection is sought. Part II can be further 
divided into (a) Sn. 224-226 (b) Sn. 227-235. (a) Sn. 226 marks the final 
stanza said to have been uttered by the Buddha according to one tradition 
quoted by the Commentator, Apare pana vadanti: adito pane eva gatha 
bhagavata vutta, sesa parittakaranasamaye Anandattherena ti. (Others 
maintain that only the five opening stanzas, viz. Sn. 222-226, were uttered 
by the Blessed One, and the rest by the Elder Ananda on the occasion of 
the paritta-recital. — Pj. I. 165). It is interesting to note that up to this 
point the Sangha is not mentioned and it is noteworthy that the only other 
references to an organised Sangha in Sn. are at 519d and 1015b (the latter 
in the late prologue to the Parayand) 22 Although there may be the possi- 
bility of some truth underlying this tradition, Sn. 222-226, by themselves, 
do not form a satisfactory unit as a sutta. On the other hand if Sn. 223 
is rejected as a late stanza, since the stanzas corresponding to it in the 
Mahavastu version occur somewhat later in the sutta (i.e. vv. 15-16; 
Mvastu. I. 294), it would be possible to infer that Sn. 227 marks the last 
stanza of a complete unit, thus partly agreeing with the tradition mentioned 
above, (b) Seven (i.e. Sn. 227-232, 235) of the nine stanzas in this group 
are devoted to describe the Sangha: and it is apparent from the over 

21. C.pNd. 2. 523 : 

22. There are other references to nagasangha at Sn; 421b and samanasangha at 
Sn. 550c in the more general sense of "multitude" as in devasangha at Sn. 680c. 

102 Pali Buddhist Review 2, 2 (1977) 

emphasis laid on the Sarigha that this section was one of the so-called 
"monastic fabrications". There are nine consecutive stanzas in Mvastu. 
(Viz. vv. 6-14: Mvastu. I. 291-294) with the refrain, idam pi sahghe ratanam 
pranltam, etc. Of the other two stanzas, Sn. 233 and 234, the former 
roughly corresponds to v. 15 in Mvastu, (1. 294) which runs, prathame, Caitrasmin 
vane pragulmd yatha puspitdgru 
vaterita te surabhim pravanti 
evamvidham dhyayino Buddha putrah 
silenupeta surabhim pravanti 
idam pi sanghe ratanam pranitam 
etena satyena suvasti bhotu 
manusyato va amanusyctto va 

The latter (Sn. 234) has no parallel in Mvastu. From this disparity it may 
be surmised that the BSK. version was based on different recension of the 
sutta which perhaps was earlier than or contemporaneous with the Pali. 
The better arrangement of the expanded version in Mvastu. perhaps 
indicates that as a sutra, it is younger than the version preserved in Sn. 
Part III which consists of the concluding stanzas is attributed neither to 
the Buddha nor to Ananda, by the Corny, and tradition, but to Sakka. 
The last two verses are mere repetitions of Sn. 236 with the Dhamma and 
Sangha substituted for the Buddha in line c. It has already been observed 
that these three stanzas show the development of a complete doxology 
in Buddhists worship. All the three stanzas are found in a condensed form 
in the concluding verse in Mvastu, (I. 295). 

On purely linguistic evidence the sutta appears old; but its contents 
and developments in ideology and doctrine show that it cannot be as old 
as the earliest suttas of Sn. Many of the archaic and dialectical forms 
found in the sutta have no exact parallels in Mvastu. This is clear evidence 
that the recension from which the sutra in Mvastu. was compiled had lost 
sight of such old forms, and found them too obscure to restore the equival- 
ent Sanskritised forms. The phrase diva ca ratto ca (Sn. 223c), though 
stereotyped, is old; and the younger Pali would prefer diva ca rattiya ca 
(both inst.) or divam ca rattim ca (both ace.) as in Mvastu. divam ca ratrim 
ca. The adverb of place huram (Sn. 224a, cp. Th 1 . 10, Sn. 486c. 470c, etc.) 
goes back to the dialect of the Brahmanas). 23 Even though the phrase 
satam pasattha (Sn. 2Tlz) is neither irregular nor particularly archaic 
Mvastu. has sada prasastd. It is quite probable that the recension that 
Mvastu. followed contained the idea sada and not satam. In Mvastu, the 

23. A. V. huruk> Brah. huras>-P. huram (with -am the standard adverbial termi- 
nation formed after the ace. sg. of nt. nouns'). Cp. hurahuram at Dh. 334, Thl . 339, etc. 
Mvastu. has parasmin in place of huram. 

Pali Buddhist Review 2, 2 (1977) 



cpd. suppayutla (Sn. 228a) is replaced by yuktayogl (cp. yogayukto 
munir Bhg. V. 6), and laddhd mudha by a totally different idea, vimuktacitta. 
Though mudha is met with in Classical Skr. it is an old form. The form 
catubbhi is historical and old (Mvastu. caturbhi). The archaic particle 
su (cp. Skr. svid) at Sn. 231b in the phrase tayas su dhamma is lost in 
Mvastu. (trayo 'sya); so is uda (cp. Vedic uta) at Sn. 232b replaced by atha. 
The enclitic no in the phrase na no samam atthi if perhaps dialectical 
(Mvostu. tarn). Similarly pronouns ye and ne at Sn. 223cd are not only 
lost in Mvastu. but the corresponding lines there, 

divam ca ratrim ca haranti vo balim 

tasmaddhi nam raksatha apramatta (Mvastu. I. 294) 

are grammatically wrong. The old Magadhl nom. sg. in -e at Sn. 233a 
cannot be traced in Mvastu. which has the plural instead. The cpd. 
paramam hitaya is an old a/w/fc-samasa which has puzzled even the Com- 
mentator who explains the nasal as metri causa, (Pj. I. 192). If that was 
so parama (-a contraction for -ayd) which would suit the context better 
and has the same metrical value is to be expected here. 

The style of the sutta is simple and the verses are quite vigorous. There 
are a few similes used e.g. the inda-khila 2!> at Sn. 229ab, the forest-grove in 
summer at Sn. 233ab, etc. Metaphor is not infrequent, e.g. khinabija at 
Sn. 235; etc. The heaping up of the attributes of the Buddha at Sn. 225a, 
234ab may be a sign of a more developed style. The language is essenti- 
ally the poetic diction. 

The metre of the poem is Tristubh, but there are numerous metrical 
irregularities such as 17 Jagatipddas, one instance of a pdda with caesura 
after the seventh syllable not being reckoned (Sn. 223a) and a contaminat- 
ed pada (Sn. 235b) enumerated by Helmer Smith. 25 

There are many points of doctrinal importance in this sutta. It is 
not possible to specify any of them as old or young, but certain trends are 
noticeable. Both old and more developed concepts lie side by side. 
Mettd has already been discussed (Sn. 223b). The categorical statement 
that the Buddha, Dhamma and the "uninterrupted samadhV are unique 
(Sn. 224c, 225c, 226c) shows signs of a developed lore. There is an 
elaborate theory almost amounting to a dogma discussed in Sn. 227-232. 
These verses are quite valuable in tracing the Arahant-ideal in Buddhism. 26 
Most of what is stated here is found in the older parts of the Canon; yet 

24. S.v. PTS. ' 'The post, stake or column of Indra, at or before the city gate; also 
alargeslabof stone let into the ground at the entrance ofa house". Pj.I. 185: Nagara- 
dvaravinivaranattham ummarabhantare affha va dasa va hatlhe pathavint khanitva akoti- 
tassa saradarumayatthambhass' etarri adhivacanar/i. 

25. Op. cit. (SnA. 683 ff)., 

26. Vide Miss I. B. Horner, Early Buddhist Theory of Man Perfected. 



Pali Buddhist Review 2, 2 (1977) 

there appears a slight departure in the method of presentation. The 
basic tenet of the ariyasaccdni occurs at Sn. 229d and 230a. The eight 
puggalas (individuals) culminating with the Arahctnt are mentioned (Sn. 
227ab) and there is a probable link with yoga in the phrase suppayutta 
(Sn. 228a. — in perfect control). It is emphasised that these puggalas will 
not enter an eighth existence (Sn. 230d). Further evidence for the develop- 
ment of the concept Arhant is to be seen in Sn. 231 where it is stated that 
the individual (belonging to one of these eight categories) has abandoned 
sakkdyaditthi "heresy of individuality", vicikiccha "perplexity" and 
silabbatapardmdsa ' 'the observance of diverse vows and ascetic practices ' '. 
These three concepts represent a somewhat developed phase. As opposed 
to vicikiccha is saddha which signifies a religious aspect rather than a 
moral relation. The "contagion" of various silas and vatas may 
perhaps refer to various types of Brahmanical and other ritual. There 
also occurs a minor dogma (not found in Mvastu.) at Sn. 23 le. If these 
two lines do not belong to a later stratum than the rest of the poem, the 
term abhifhdndni may also indicate general lateness. The six grievous 
offences include the five anantarika kammas and dnnasatthdr'uddesa 
(pesiting another teacher — cp. micchdditthi). The latter was probably 
added to the earlier list of five with the arising of a growing rivalry between 
the Sangha and the members of other sects.' Further attributes of the 
Arahant occur at Sn. 232. Although Arahants are mentioned in the earlier 
part of the Canon and arahatta is a familiar concept, the sutta definitely 
reveals a development in the theory of the Arahant. The centre of gravity 
has already shifted from the muni to the "perfect being". This is the 
outcome of a widespread monastic organisation as opposed to the "lonely 
wanderers" of the older ballads. 

External evidence — A parallel version of this sutta occurs at Mvastu. 
I. 290 ff. As stated earlier the two versions in Pali and BSK. may be 
traced to a common source with different recensions rather than one being 
based on the other. The evidence discussed above shows that the sutra 
in Mvastu. is relatively younger than the Pali. Though doctrinally the 
Pali version depicts comparatively developed phase of Buddhism, linguistic 
and external evidence debars one from assigning a very late date. 27 
A passage found at Divy. 340 throws considerable light on both versions 
of the sutta. It runs; Ayusmad Sangharaksiteha Ndgaropamam siitram 
upaniksiptam gdthdm ca bhdsate, 

Yaniha bhutdni samdgatdni 
sthldni bhdmydm athavdntarlkse 
kurantu maitrim satatam prajdsu. 
diva ca rdtrau ca carantu dhdrmam 
27. The Mvastu. vortion, however, to decidedly later than the Pali poem. 

Pali Buddhist Review 2, 2 (1977) 


The Nagarupama Sutta at A. IV. 106 IT. has no connection whatsoever 
with the verse quoted, nor with Mvastu. I. 290 ff. The stanza may be 
compared with v. 15 of the sutra in Mvastu. (I. 294), lines ab of v 2 
(Mvastu. I 290) and Sn. 222ab, 223bc. The same idea is found at Brh. 1. 4, 29 
and IV. 3, 43. It is not very probable that Divy. 340 quotes from Mvastu. 
or Pali. On the other hand, it may perhaps be attributed to some source 
which may have been connected with the original version of the sutta. 
The other probability is that the opening verses of the Ratana Sutta and 
the corresponding gdthds of Mvastu. have drawn upon this stanza, which 
may have originally belonged to some other section which in all probability 
was the Nagafopama Sutra mentioned in Divy. (and not the sutta bearing 
that name at A. IV. 106 ff.) This sutta seems to have consisted of general 
advice given in the form of an address made to the bhutas, for, the second 
line inculcates the practice of maitri and dharmacarya. There is ano.ther 
reference to the Triad of Ratanas at Divy. 481, in the form of a salutation 
(namo ratnatrayayd), which merely shows that a conception of such a 
triad was familiar to the editor of that section of Divy. 

Indirect evidence— Internal evidence and all available external evidence 
show that the sutta is comparatively late. It is also found to be decidely 
later than the Aft/m-Ballads of Sn. It has been observed earlier that 
subsequent additions are normally made to canonical works by appending 
them either at the head of a section or at the end of it. The Ratana Sutta 
is clearly an addition made to the Culla Vagga, after a vagga as such had 
been formed. Evidence of this nature is not helpful in determining dates 
ofsuttas, but on the other hand, it is an invaluable source of information 
in tracing the growth of the various works. It is also useful as a con- 
firmatory test to what has already been discovered from other sources. 

Conclusion— The supposition that the opening stanzas of the sutta in 
Sn. are based on a sutta which is now lost leads to the natural inference 
that the Ratana Sutta is of a rather composite nature, being built up of 
various elements at different periods. Though there is no conclusive 
evidence forthcoming to establish the tradition, a comparison of the two 
versions of the sutta in Pali and BSk and an analysis of its contents have 
shown that this is true to some extent. Generally speaking, parts I and 
III (supra) have been observed to be on a different level from the rest of 
the poem; and parts Ha and lib on two separate levels. Doctrinal 
evidence has shown that the two additional padas Sn.23lgh are consider- 
ably late; and similarly the short stanza Sn. 234 which gives a categorical 
list of attributes of the Buddha appears to be later than the longer stanzas 
occurring immediately before and after it. 


,w fall Buddhist Review 2, 3 (1977) 

92. By the destruction of ignorance, the desire of existence disappears 
and by the suppression of the desire of existence, rebirth ceases and 
by the extinction of rebirth, grief, infirmity and dissolution cease- as 
if the light of a candle is put out. 

93. Beings! It is said by our Lord Sakya Sinha, that he who in this 
world examines the constitution of the Buddha's Doctrine of Enlighten- 
ment sees the Buddha himself. Endeavour to comprehend the Teacher 
of the Three Worlds,, and his holy doctrine: it is the invariable practice 
ot the virtuous to do so. 

94. Beings! (To be saved) it is absolutely necessary, by the hook of 
wisdom, to pull out the principal causes of sin (avarice, malice and 
ignorance) which are unpleasant and equal to a sharp thorn piercing 
the heart, and which lead to ruin and every mode of evil. 

95. As Mount Meru stands unshaken, before a strong wind so the 
heart, pure and free from the attraction of existence, is always unshaken 
by the eight vicissitudes (lokadhamma) of the world and the five desires 
(kilesas) in their various and complete stages. 

t> 9 fju Be T ! A1W3yS giw y ° Ur tjme for the S° od of others, as the 
Buddha, the Supreme Lord, who sailed across the deep ocean of ten 
noble attributes (paramitas), regardless of the horrors of metamorphosis 
and who destroyed the flame of ignorance and by self-exertion gained 
a lull knowledge of all that should be known. 

97 Suspending the attainment of Nirvana, at a time when he was 
not far from it, Sakya Muni passed many a dreadful birth on account 
of his sympathetic feelings toward others. So do I give you instructions 
chiefly with the object of doing good to others. 

98^ It is impossible that any one of sound understanding should 
wander away from the noble path of the Law, after attaining The most 
difficult object of attainment, the human for, and after gaining all the 
varieties of wealth and luxury and a comprehensive knowledge of the 
doctrine capable of giving a death-blow to the desires of existence 


N. A. Jayawickrama 



ALTHOUGH the three Suttas, Hemavata, Alavaka and Suciloma are 
fundamentally similar in that they are riddle-poems resembling the 
Yaksa Prasnas of the Mahabharata, the Hemavata Sutta demands special 
attention on account of its extraordinary length and the difference it 
bears to the other two in details. Unlike the other two suttas it contains 
no prose introduction and its principal characters Satagira and Hema- 
vata are represented as friendly beings whereas the two yakkas Alavaka 
and Suciloma are no more than mere demons. 1 All the three suttas are 
dialogue-ballads, but the dialogue consists of only one question and an 
answer to it in the Suciloma Sutta, while there are only two characters 
in the Alavaka Sutta. The dramatic element is quite pronounced in the 
Hemavata Sutta, and the Alavaka Sutta is not devoid of it. 


Hemavata Sutta 

The sutta begins as a conversation between Satagira. "the dweller on 
the Sata Hill in Majjhimadesa", and Hemavata, "the Hime'ayan 
Sprite". 2 When the former succeeds in convincing the latter of the 
virtues of the Buddha, they visit him, and Hemavata who plays the 
role of the questioner throughout the poem asks the Buddha questions. 
The two yakkhas are delighted with his answers: they extol him and, 
along with their followers, seek his refuge. 

The sutta may be divided into three parts, viz: (i) Sn. 153-167, the 
dialogue between the two yakkhas, (2) Sn. 168-175, the dialogue between 
Hemavata and the Buddha, (3) Sn. 176-180, the conclusion which con- 
sists of an exaltation of the Buddha. 

Part I. When Satagira invites Hemavata to visit Gotama, the latter 
asks him whether Gotama possesses various qualities, which Satagira in 
his answers affirms. All the virtues of the Buddha which are enumerated 
in this dialogue may equally well be attributed to any sage. Even the 

1. Ajavaka: "of the forest", from atavi, forest; Suciloma: "needle-haired". 

2. SnA. 197. W. Stede suggests that Satagira may be a variant for Satagila, a second- 
ary form of Satagila "swallowing a hundred". He further suggests that Nalagiri. 
stands for Naragila. 

142 Pali Buddhist Review 2, 3 (1977) 

few stanzas which are meant to describe the personal attributes (Sn. 
165-167) lay emphasis on his ascetic life and not his person. In fact 
the whole poem emphasises the conduct (cariyd) and the spiritual attain- 
ments of the Buddha (the word used is citta— the mind). The Buddha 
at most here is a perfect muni and is not spoken of in the grandiloquent 
terms that usually accompany a developed phase of Buddhism. 

Part II. Hemavata asks a question which may be interpreted as being 
of consmological or cosmogonical import, as it vaguely touches upon, 
the adi and anta of the universe (Sn. 168). The Buddha gives a cryptic 
answer from a teleological standpoint, but does not enumerate nor 
specify what groups of six he alludes to (Sn. 169). In answer to the next 
question in which the yakkha shows that he has understood the allusions 
to the ayatanas the Buddha mentions the pleasures of the five senses 
and the mind as upadana (grasping) as the sixth and states that their 
abandonment leads to emancipation from misery (Sn. 171-172). The 
yakkha then asks the nature of the person who crosses the Flood (ogha 
— Sn. 173) and the Buddha describes the virtuous sage who has gained 
spiritual attainments and "does not sink into the deep". (Sn. 174). 

Part III. The two yakkhas praise the Buddha (Sn. 176-177) and compli- 
ment themselves for having taken the opportunity of visiting him (Sn. 
178). They along with their 1,000 followers seek his refuge (Sn. 179) 
and make a solemn pledge to honour the Buddha and the Dhamma 
(no Sahgha is mentioned) in all their wanderings (Sn. 180). Even here 
the epithets used of the Buddha are those of the perfect sage. 


The language of the sutta is the standard poetical Pali. There are many 
poetical expressions as divya ratti (Sn. 153b), anomanamam (Sn. 153c, 
177a), samsuddhacarano (Sn. 162b, 163b), khlnavyappatho (Sn. 158b: 
Comy, khino vacaya patho cp. Sn. 1076d: vadapatha) and vyappatha (Sn. 
163Ab, 163Bb, 164b, cp. vyappathayo Sn. 961a). The frequent use of 
the interrogative particle kacci expressing doubt (18 times) and the dis- 
junctive indeclinable atho (7 times) is very striking. Though the indie. 
3 sg. aha (Sn. 158c) is the normal Aor. (pf.) 3 sg. it is used here in the 
present tense as in several other old suttas. 

The syntax if the verses is generally straightforward, but there are a 

few instances of ellipsis; e.g. Sn. 168, 169. The Corny, correctly takes 

kismim at Sn. 168abd 3 as locative in one or other of its basic meanings 

3. SnA. 21off. kismim at Sn. 168a is explained as: bhavena bhavalakkhane bhumma- 
vacanam (being a condition, it is a locative denoting condition), at 168b: adhikarai.xatthe 
bhummavacanai>i (locative in the sense of relation— in time) and at 168d: bhavena bha- 
valakkltana-karanatthesu bhummavacanani (being a condition it is a locative characteris- 
ing condition and denoting cause). 

Pali Buddhist Review 2, 3 (1977) 


and equates it in the first pada to kismim uppanne and in the fourth 
pada to kismim sati. The explanation of chassu in the corresponding 
stanza (Sn. 169) is similar. 

As regards style the sutta stands out as a highly dramatic piece. Like 
many other similar ballads this too was probably sung on suitable occa- 
sions, three different reciters singing the stanzas ascribed to each of the 
characters. The poems appears to be divided into two separate scenes, 
for the dialogue between the two yakkhas takes place at one place and 
their conversation with the Buddha at a different place. Except for a 
few occasional poetical flashes the style invites no comments. There 
are two highly descriptive passages in the sutta, viz. Sn. 165-167, 176-180. 
The simile slham v'ekacaram (Sn. 166a) is reminiscent of numerous 
others of lonely wanderers (vide Khaggavisana Sutta). The repetition of 
the same words at Sn. 163A., 163B and 164 is due to a subsequent expan- 
sion probably effected in Burma. 4 

The group of six indicating a set of phenomena need not necessarily 
be late, and the contents show that the reference is to the ayatanas^ an 
early concept in Buddhism. The grouping of epithets at Sn. 167, 176, 
177 is to be generally regarded as a sign of lateness, but the absence of 
any indication of a developed Buddhology attributing supernormal 
qualities shows that these verses may still be old. There are no indica- 
tions of these verses being later than the rest of the poem. 

Metre. The Anusiubh slokas of the poem are interrupted by two 
stanzas in Tristubh. (Sn. 176, 177). The repertory phrases (iti Satagiro 
yakkho, etc.) which the Corny, attributes to the sangitikara (SnA. 193) 
do not fall within, the metre. The break in the metre may perhaps indi- 
cate that the two stanzas in Tristubh were borrowed from elsewhere; but 
the full stanza Sn. 176 has not yet been traced to any other work, though 
three of its individual padas are seen to occur frequently in other metrical 
works. 5 On the other hand Sn. 177 occurs at S.I. 38 (cp. Sn. 153). Yet, 
the composite nature of the Sagathaka Vagga of the Samyutta does not 
warrant the inference that Sn. has borrowed this from there. There is 
no doubt that these st&ozas are old. The fact that they are written in 
the historically older Tristubh is further proof of their antiquity. How- 
ever, it may still be probable that these two stanzas were interpolated 
from an earlier source. There are also a few metrical irregularities in 
the poem. Sn. 153 is in mixed Anustubh and Vaitaliya (the first pada in 
Vaitaliya). Sn. 554a, 155a, contain nine syllables each instead of eight. 

4. Only Burmese Mss. and SnA. accept these two additional stanzas. 

5. Sn. 176a: Th. I, 372: Sn. I76b: Vin. I, 36, Sn. 1059b, 1091d; Sn. 176c: 
196, S.I. 4. 50, 51, Dh. 90 cp. Sn. 472c, 501 b. 

D. Ill, 


Pali Buddhist Review 2, 3 (1977) 

There are "even " padas at Sn. 156a, 157a, 158a, 159a, and the caesura 
is not marked in Sn. 176d and 177d. 6 

Doctrinal Developments. The whole poem reflects a period when the 
Af««/-ideal was the vogue in Buddhism. The emphasis of the poem on 
the conduct and mental discipline of the Buddha, the reference made 
to him as the ideal sage, the simple conception of the Buddha as opposed 
to what may be seen in later works, and the simplicity of the ideas in 
contrast with the dogmatism of the latter all point to the early date of 
the poem. The sutta is untarnished by any doctrinal colouring of a 
speculative nature. 

In their application none of the terms shows any distinct growth. In 
supanihita (Sn. 154a, 155a, cp. 163Aa, 163Ba, 164a and the term padhana) 
is found an idea parallel to yoga though the word itself does not occur 
here . Again, the yogin can be implied from Sn. 156d, 157d (jhanam na 
rihcati, also cp. terms satipatthana and bhavana). These ideas are old 
and were the common knowledge of all schools of contemporary Indian 
thought. Sn. 154b, 155b, 156b, 157b indirectly convey the idea of metta. 
The concept of moha (Sn. 160c, 161c) may be compared with moha in 
the Bhagavadgita (4, 35; 14, 13; 17; 22, etc.). It is essentially a state of 
mind unlike maya which is more in the nature of a cosmic (or' metaphysi- 
cal) state found in association with the empirical world. The term dibba 
patha has already been commented on. 7 The occurrence of the groups 
of six at Sn. 169 indicates the early classification of the ayatanas. They 
are mentioned later under the panca kamaguna and the mind (Sn. 171ab). 
The didactic element of the sutta is best judged fromSW. 174-175— two 
stanzas important from a teleological standpoint. The terms ogha and 
annava have been discussed elsewhere. None of the other terms that 
occur in this sutta calls for particular attention as they are used in all 
stages of the language. 

Judging from internal evidence the sutta appears old. Its language 
shows no signs of lateness while there are no special forms which may 
be classed as very old. Its diction is the early poetical expression. The 
numerous padas and stanzas of the second half of the poem (Sn. 163 A, ff., 
i.e. at the end of the dialogue between the two yakkhas) which are in 
common with other metrical works 8 suggests that that section of the 
poem has borrowed freely from an existing floating metrical literature. 
External evidence will be discussed after dealing with the internal evidence 
from the other two suttas. 

6. Helmer Smith, Metres of the Sutta Nipata, SnA II 3 

7. PBR I, 3. 

8. Vide E. M. Hate, Woven Cadences, p. 189 and Otto Franke, ZDMG. 

Pali Buddhist Review 2, 3 (1977) 


Ajavaka and Suciloma Suttas 

The position of the Alavaka Sutta immediately after the Hemavata 
Sutta points to an attempt at an arrangement of suttas according to 
subject-matter. This has already been noticed in the remarks on the 
Vasala Sutta. 9 However, there is no consistency in this matter, for the 
Suciloma Sutta is placed in the next (Culld) vagga, after the Mahgala 
Sutta. It is also significant that the four suttas (beginning with Fara- 
bhava) preceding Alavaka are parittas. Both the Suciloma and Alavaka 
Suttas contain a prose introduction in which the two yakkhas are seen 
to intimidate the Buddha with identical threats. The Buddha's answer 
in. both instances is the same. The first half of the introduction is some- 
what different in the two suttas, and there is another yakkha, Khara, 
mentioned in the Suciloma Sutta. In both instances the sutta proper 
begins with a question in verse, after the prelude in prose. The prose 
of these suttas is the canonical idiom, and therefore represents a later 
phase of Pali than the gathas. 10 The Yakkha Samyutta (S.I, 206-215) 
contains both these suttas in identical words. 11 It is quite probable that 
that the prose introductions were appended to the gathas during the time 
of the compilation of Sn. as an anthology, and that they were taken 
from, the legends in the Yakkha Samyutta though both works are depen- 
dent on an earlier tradition for the gathas. In spite of the fact that Miln. 
36 attributes Sn. 184 to Samyutta, both Yakkha Samyutta and Sn. are 
collections made from earlier existing material. 

The Alavaka Sutta 12 contains a series of questions and answers (Sn. 
181-190) followed by an epilogue in verse (Sn. 191-192) whereas the Suci- 
loma Sutta ends with the Buddha's answer to the question at Sn. 190. 
An examination of the former shows that Sn. 190 forms a suitable conclu- 
sion to the sutta when the line, so 'ham ajja pajanami yo attho sampara- 
yiko indicates the questioner's complete satisfaction with the answers 
he has received. The tone of the epilogue appears totally different from 
that of the rest of the poem, and resembles that of some of the concluding 
verses of the Thera-gathas or the later Apaddnas. The only connection 
of the poem with the Alavaka-legend is in the phrase, Alavim agama 
at Sn. 191b. The fact that this sutta is perhaps later than the preceding 
stanzas is also suggested by the statement, yattha dinnam mahapphalam 
(line d) which stands at a much lower level than the previous statement 

9. PBR 2, 2, p. 88. 

10. Vide PBR. 

11. Ajavaka Sutta at S.I, 213-215; Suciloma at S.I, 207-208. 

12. The events connected with the Alavaka Sutta are placed in the sixteenth year 
after Enlightenment — E. J. Thomas, Life of the Buddha, p. 1 19. 


Pali Buddhist Review 2, 3 (1977) 

at Sn. 190cd. It appears quite irrelevant that the yakkha should realize 
that the Buddha is an ideal punhakkhetta, when he should be thankful 
for the exposition of his questions. Sn. 192 occurs at Ap. 6, 152; 415, 17 
and various commentaries; and its tone appears decidedly late. It is 
highly probable that Sn. 190 formedthe original conclusion of the gathas 
and that the epilogue was a later addition concurrent with the identifi- " 
cation of these gathas with the Alavaka-legend. 

The two opening stanzas of the Suciloma Sutta (Sn. 270-271) consist 
of a question and an answer. Like the two opening stanzas of the 
Alavaka Sutta they are Tristubhs; the first two stanzas of Hemavata's 
dialogue with the Buddha are also mTrisfubh (Sn. 168-169). Questions 
and answers of this nature are found in a section of the Devata Samyutta 
(S.I, 36-45); and further, two of the above passages occur there: viz: Sn. 
168-169 at S.I, 41 and Sn. 181-182 at S.l, 42. The occurrence of these 
stanzas in the Samyutta, independently of the rest of the respective poems 
suggests the existence of a set of riddles dealing with Buddhist topics 
prior to their being incorporated in longer poems. 13 Unlike the Alavaka 
Sutta, the Suciloma Sutta seems to have been built upon one such riddle 
though Sn. 270-271 have not been traced as an independent piece. The 
third stanza in Anustubh is an explanatory verse on the answer to the 
riddle. The general appearance of a sutta is given by the addition of 
the concluding stanza. It is obvious that many Pali poems have incorpo- 
rated earlier existing material, but Suciloma Sutta appears totally different 
as it is evident that it is built on the framework of the riddle. The four 
gathas as a whole, appear old, but on the basis of this argument Sn. 
270-271 are older than the other two. 


The language of the suttas is considerably old; and the Alavaka Sutta 
preserves many dialectical as well as old historical forms; The phrase 
mittani ganthati (Sn. 185d, 187d) preserves the historical gender of mitta, 
although in Pali the word is masculine. The idiom itself is perhaps Vedic 
or post-Vedic rather than Classical Sanskrit or Pali. The word sussusa 
is to be taken as a shorter inst. sg. (Vedic) and not as a contraction of 
the Pali sussusaya. There is a Vedic ppr. in saddahano (cp. srad-dadhana 
— Sn. 186a); there are dialectical forms as ingha and bhiyyo (Sn. 189 — also 
in prose). The particle of interrogation su is frequently added to the 
interrogative pronouns to emphasise the question * as is characteristic of 

13. The tradition preserved in the Devata Samyutta may perhaps be synchronous 
with a floating riddle literature which was the predecessor of Sanskrit riddle poetry. Also 
vide Winternitz, History of Indian Literature, I 352 on old riddle poetry. 

Pali Buddhist Review 2, 3 (1977) 147' 

old gatha-?3i\\ (cp. Epic sma). Other old forms are: agent noun utthata 
(Sn. 187b) and imp. 2 sg. Atp. pucchassu (Sn. 189a). There are hardly 
any peculiar forms in the Suciloma Sutta. The forms kutonidana, kutoja 
(Sn. 270), itoniddna, itoja (Sn. 271) are common poetical forms. The 
reading dhamkam (crow) should be preferred to vankam (see also SnA. 

Style. Both suttas are dialogue-ballads, but the dialogue is more pro- 
nounced in the Alavaka Sutta. The moral truths are stated expressively 
in a series of questions and answers in clear and simple language (cp- 
the opening stanzas of Bhg. VIII). The sutta lends itself to easy dramati- 
zation on account of its being well punctuated by the words of the two 
interlocutors. The concluding stanza (Sn. 192) like Sn. 179-180, enhances 
its dramatic effect. The Suciloma Sutta, though short, is more ornate 
than the other. It contains three similes viz. Sn. 270d-271d, 272b, 272d; 
and the stanza Sn. 271 is rather cryp':ic. 

Metre: The break in metre in the two poems has already been noted 
(supra). Although Tristubh is historically older than Anustubh sloka, it 
in itself provides no useful data. As observed earlier, the occurrence of 
one of these Tristubh passages (Sn. 181-182) at S.I, 42 and a similar 
passage (Sn. 168-169, though in Anustubh sloka) at S.I, 41 along with 
many other similar questions and answers, in the Devata Samyutta, may 
presuppose the early existence of a traditional riddle-literature indepen- 
dent of longer suttas. The disparity in metre is therefore due to the 
fact that some of these passages are either incorporated in, or utilised 
to build up (as in the case of the Suciloma Sutta) longer poems. The 
two opening stanzas of both suttas, thus appear to have belonged to 
an altogether different stratum from the rest of the two poems. Metrical 
irregularities are almost absent in the Anustubh slokas (Sn. 184-192,271). 
There are two instances of even quarters at Sn. 186a and 187a. Of the 
Tristubh stanzas (Sn. 181-182, 270-271, 273) Sn. 270b-271b are irregular. 
Besides containing anacrusis, the caesura after the seventh syllable is not 
reckoned in them. 14 . There are two instances of metrical lengthening, 
viz. avahati (Sn. 181b-182b) and larati (Sn. 183a-184a). The lengthening 
in su'dha (Sn. 182a) is due to sandhi. 

Doctrinal Evidence: Notwithstanding a few words with a semi — or 
quasi-technical significance, such as ogha, annava, appamada, viriya and 
panna (Alavaka), raga and dosa (Suciloma), the two suttas are marked 
by a total lack of metaphysical thought. The simplicity of ideas and 
the emphasis laid on saddha shed some light on the antiquity of the 
Alavaka Sutta. The popular character of the whole poem is also seen 

~s ■ 

14. Helmer Smith, ibid. 

148 Pali Buddhist Review 2, 3 (1977) 

from the occurrence of the word idha (here, on earth) in saddWidha (Sn. 
182a). The popular teachings embodied in the sutta (in Sn. 188-189 
which state the four fundamental qualities sacca, dhamma — or dama, 
dhiti — or khanti, and caga, requisite of a devout householder) are reminis- 
cent of the Dharmasdstras and other allied branches of Sanskrit literature. 
The practice of one's dharma, truthfulness, discriminate living, diligence, 
energy, wisdom, desire to learn, doing what is correct, tenacity and 
perseverance all contribute to help a person to reap the benefits of 
learning, fame, wealth, acquisition of friends, etc. These teachings were 
accepted by all schools alike as Sn. 189 points out. The answer to the 
important question of teleological significance (Sn. 183) emphasises the 
role saddha plays in popular Buddhism. As the contents show, this 
sutta dealing with popular teachings appears old and seems to have 
drawn freely from the fund of old Indian knowledge. The Suciloma 
Sutta which deals with detachment from passions is particularly more 
Buddhistic than the other. Though the sutta itself is too brief to draw any 
inference on doctrinal grounds, it has the general appearance of an old 
piece (specially Sn. 270-271). The poem as a whole may not be of very 
great antiquity, but the two opening verses, for reasons discussed above, 
are at least as old as the Hemavata and Alavaka Suttas. 


The Yakkha-legend (Alavaka) 

The following observations on the yakkha-hgend shed some light or 
our suttas. The Commentary connects the story of Alavaka with that 
of Satagira and Hemavata (SnA. 221ff,). When the two yakkhas Satagira 
and Hemavata were on their way to Jetavana in order to pay their 
respects to be Buddha before proceeding to the assembly of the yakkhas, 
they found it impossible to pass over Alavaka 's abode. On investigating 
the cause they found the Buddha there, saluted him, listened to the 
dhamma and continued their flight to their final destination. A similar 
story is found at UdA. 64 when they passed the abode of Ajapalaka- 
yakkha. In this story is seen a popular attempt, however late it may be, 
to link up the various yakkhas with one another. The legend of Alavaka 
occurs in similar words at SA . 317ff. in the commentary on the sutta in 
the Yakkha Samyutta. A summary of the same legend occurs at A A. 
389ff. in the comments on Hatthaka Alavaka, a prince who is said to 
have been saved from the yakkha' s hands. 15 The story itself has much 
in common with the circle of legends grouped by Watanabe 16 under the 

15. In the list of etad-aggas at A.I, 26 he is called the most pre-eminent of those who 
cherish the assembly with the four sangaha vatlhu (bases of generosity). 

16. JPTS., 1909-1910, pp.240ff. 


Pali Buddhist Review 2, 3 (1977) 149 

title Kalmasapada Stories. The essential links are: (1) the man eating 
yakkha, (2) the captured king who obtains his freedom by promising to 
provide the yakkha with food, and the sanctity of that promise, and 
(3) the conversion of the yakkha. This similarity to the Kalmasapada 
group is sufficient proof of the antiquity of the Alavaka-legend. 17 


External Evidence 

There are sufficient references in the Canon to most of the yakkhas 
who are represented as taking part in these dialogues. Hemavata and 
Satagira are included in the list of great yakkhas whose protection should 
be sought when troubled by other yakkhas. 1 *' Their names represent a 
class of yakkhas (probably their followers) in the Mahasamaya Sutta. 
Cha sahassa Hemavata yakkha. . . and Satagirati sahassa yakkha (D. II, 
256) may be contrasted with Ime dasasata yakkha at Sn. 179 which 
speaks of a following of 1,000 instead of a total of 7,000. 

The occurrence of the Alavaka and Suciloma Suttas in identical words, 
in the Yakkha Samyutta has already been mentioned. Besides this the 
statue of Suciloma which is found at the Bharhut Stupa along, with 
those of many other yakkhas is a fair land-mark indicating the early 
acquaintance with all these yakkhas. 19 There is no doubt whatsoever 
that all these yakkhas were known, as seen from the references in the 
Pali Canon, long before the time of Bharhut, yet the Stupa itself'is 
helpful in determining the lower limit of the date of these legends. 

No remarks need be made here on the concept yakkha, the part yakkhas 
play in Indian literature or the origin of the yakkha-culi. These questions 
have been exhaustively dealt with by scholars like A. K. Coomaraswamy 
(Yaksas), W. Stede (Gespenster geschichten des Peta Vatthu; and s.v. 
P.T.S.), Dela Vallee Poussin (Indo-Europeens et Indo-Iraniens; Ulnde 
jusque vers 300 av. J.C.), O. H. de A. Wijesekera (U.C.R. I, 2) etc. It 
is also of no importance to investigate further the Commentarial accounts 
linking up the legends of various yakkhas. The question to be solved 
is how these yakkhas came to be associated with the gathas in Sn. As 
hinted at earlier, the suttas may be analysed roughly into four different 
strata: (1) riddles which perhaps preceded the rest of the suttas (Sn. 

17. Malalasekera, D.P.P.N. (s.v.). 

18. D. Ill, 204-205; Atanafiya Sutta. Cp. the statement at SnA. 197, Hemavata- 
Satagira atthavisati ydkkha-senapatinairt abbhantara mahanubhava yakkharajano 

19. Cunningham, The Stupa of Bharhut. He assigns the date of the Stupa as 
250-200 B.C. (p. 14): The inscription under Suciloma is Suciloma-yakho (p. 136 and plate 
XXII). Also vide Coomaraswamy, Yaksas, T. p. 5. 

150 Pali Buddhist Review 2, 3 (1977) 

168-169, 181-182, 270-271), (2) the poems incorporating the riddles (Sn. 
153-167, 168-178, 181-190), (3) The concluding stanzas of the suttas (Sn. 
179-180, 191-192), or the comment in verse Sn. 272 and the conclusion 
Sn. 273 and (4) the prose. 

The fact that no reference whatsoever is v made to yakkhas in parts 
1 and 2 (except in the repertory phrases which the Commentator 
attributes to the sahgitikara — SnA. 193) may probably indicate that these 
poems were at one stage quite independent of yakkhas. Internal evidence 
has shown that the concluding stanzas bear signs of lateness. The identifi- 
cation of these suttas with the various yakkhas is simultaneous with the 
inclusion of these stanzas in the poems. The number of followers of 
Hemavata and Satagira mentioned at Sn. 179 being less than that in 
the Mahasamaya Sutta one is prevented from assigning a very late date 
to the concluding stanzas, for presumably the number mentioned in the 
latter points to a more developed legend. The introductions in typical 
canonical prose are definitely of a much later stratum than the gathas. 
From the evidence at hand it is not possible to determine specifically 
when these successive additions were made. The internal evidence is 
very convincing that the gathas in parts 1 and 2 belong to an older stage 
than the rest of the respective suttas. It is quite probable that the prose 
(like many other prose passages in Sn.) was taken from outside (in this 
case the Samyutta) at the time of the final collation of Sutta Nipata as 
an anthology. 



Dhaniya Sutta 

THE Dhaniya Sutta consists chiefly of a dialogue between the rich 
herdsman Dhaniya and the Buddha. A thud interlocutor (Mara) appears 
towards the end of the sutta, and the narrator himself interrupts the 
dialogue by describing the scene at Sn. 30. 20 The recitation of the ballads 
may have proceeded on a line quite similar to the dialogue in the (later) 
medieval European Miracle Play. The dramatic element predominates 
in the sutta and the possibility of different reciters singing the respective 
stanzas attributed to the various characters has been suggested earlier. 21 
It may be said that more definite signs of dramatic representation are 
seen in the sudden appsarance of Mara voicing the popular opinion 
(Sn. 33, 34). At the same time doubts may be cast on the genuineness 

20. Vide SnA. 42. 

21. PBR 1, 2, p. 90. 



Pali Buddhist Review 2, 3 (1977) 151 

of these two stanzas; for, firstly Sn. 32 appears to be a suitable conclu- 
sion to the poem when the herdsman and his wife pledge to practise 
the holy life; secondly, the recurrence of these two stanzas in totally 
different contexts at S.I., 6 and 107-108 22 suggests that they belonged to 
a stock of traditional twin-verses, best represented by the Sagathaka 
Vagga of the Samyutta and the Yamaka Vagga of the Dhammapada; 
and finally, the greater metrical perfection indicating them as distinct 
from the rest of the stanzas may also point to a difference in the date 
of composition. Although this is no conclusive proof, it may be 
surmised with some degree of accuracy that these two stanzas were a 
subsequent interpolation. 

The sutta is mainly a poetical duel between the two chief interlocutors, 
' 'the one rejoicing in his worldly security and the other in his religious 
belief ". rf In the alternating stanzas which are highly artistic the Buddha 
invariably uses the very words of the herdsman either to express the 
exact opposite or to give a new value to them. Sometimes even the same 
sounds are reproduced with identical metrical value but expressing 
something totally different, e.g. Sn. 18a, 19a pakkoduno duddhakhiro: 
akkodhano vigatakhilo. The opposite ideas are expressed in Sn. 18c, 
19c with a different connotation for the words repeated viz. channa 
kuti ahito gini : vivata kuti nibbuto gini.™ Sometimes the words in the 
corresponding stanzas differ considerably, though in each case the speaker 
makes a statement to illustrate his point of view, e.g. Sn. 20, 21. The 
contrast is shown only in the topics discussed in Sn. 22, 23, i.e. gopi and 
cittam respectively. The Buddha is seen playing on the- word bhata 
when Dhaniya says that he is self-supporting (Sn. 24, 25). The next 
stanza of the Bhagava is a mere negation of the herdsman's statement. 
When Dhaniya speaks of tethering his animals the Buddha declares 
that he has broken all bonds (bandhandni) and will not seek birth again. 
Another pair of alternating stanzas concludes the poem when Buddha 
categorically denies the tempting words of Mara (Sn. 33, 34). 


Language and Syntax 

Proceeding on to an examination of the internal evidence, language calls 
for attention first. The sutta contains many archaic and poetical forms. 

22. In the former instance the stanza is attributed to a devata, and in the latter to 

23. Fausboll: Translation of the Sutta NipSla, SBE. Vol. X. p. 3. 

24. The Commentary (Sn.A. 31) says that kuti refers to the body; kuti'ti attabhavo. 
kayo tipi, guha ti pi (Sn. 112),deho ti pi, sandeho li pi (Thl. 20 Dh. 148) nava ti pi (Dh. 
369), ratho ti pi (S. IV. 292), dhajo ti pi, vammiko li pi (M.I. 144), kuti ti pi, kuikd ti pi 
(Thl, 1, etc.) vuccati. 

152 Pali Buddhist Review 2, 3 (1977) 

There occur several special poetical compounds aspakkodano,duddhakhiro* 
(Sn. 18a,) akkodhano, vigatakhilo (Sn. 19a), and samanavaso (Sn. !8b, 
cp. samaniya, Sn. 24b). The lengthening in patthayasi in the refrain is' 
partly archaic and partly metrical. Besides this there are other archaic 
verbal forms such as vijjare (Sn. 20a, cp. hannare, bhasare, etc. and Vedic 
sire and lsire> Pkr. -ire> P. -are, vide Geiger, 122.2); sakkhinti (Sn. 
28c) historical future homy/ sak, sak-sya-> *sakkhya-> sakkhi- with 
samprasarana; and other historical forms as upessam (future, Sn. 29c), 
pavassi (Aor., Sn. 30b), abhasatha (Sn. 30d), addasama (Sn. 31b), cwa- 
mawe (Sn. 32b) and bhavamase (Sn. 32d) which call for no particular 
attention. An interesting nominal form is gini (Sn. 18c, 19c) which is 
dialectical as well as archaic and perhaps poetic. This form probably 
comes from a dialectical stratum. The initial vowel has dropped off due 
to loss of accent. (Vedic agni>P. aggi/ agginij gini, cp. dtma>atta/ 
atuma/tuma). 25 Other noteworthy forms are: samvasiya at Sn. ,22b, 
(diaeretic, cp. samaniya, Sn. 24b), nibbitthena at Sn. 25b (adverbial inst.) 
and the sandhi usabho-r-iva with the introduction of the pseudo-organic-r; 
Skr, vrsabha-iva>P. usabho-iva, the deleted visargct is restored to bridge 
the hiatus. The word deva, in the refrain is used in the popular sense of 
cloud and the p.p. nibbuto (Sn. 19c) in its original meaning. 

The syntax of the poem also shows that its language belongs to an 
early stratum of Pali. The free use of the genitive with the verb sunati 
(e.g. tassa na sunami k ihci papcm—l hear no evil of her— Sn. 22c, cp. 
Sn. 24c, and sutva devassa wttsato— hearing it rain— Sn. 30c) is an early 
construction. The predication of a plural subject with atthi is a poetic 
usage (Sn. 26, 27). 26 The syntax of brahmacariyam Sugate caramase. 
(Let us practise the holy life under the Sugata— Sn. 32b) also strikes as 
belonging to old Pali. The nominal prefix anw-in anutire (Sn. 18b, 1 9b) 
and the phrase, tinno paragato expressing the early Buddhist concept 
of "crossing over to the Beyond" are old. 


A few remarks on the style of the sutta have already been made in the 
introduction. The poem stands out as a product of great literary skill 
and high poetic genius on the part of the composer. With a skilful use 
of words effecting as little change as possible in the alternating stanzas 
the author has been successful in illustrating the different points of view 
of the two interlocutors. The choice of words and expressions is very 
apt and lends a majestic air to the whole poem. The ballad is no mere 
versification unlike the artificial poetry of the late Ceylon Chronicles. 


Tuma <Vedic tman a by-form of atmdn. 

It is a petrified form even in canonical prose. 

Pali Buddhist Review 2, 3 (J 977) 153 

All the stanzas are very descriptive and the words of the herdsman paint 
a beautiful picture of a pastoral scene. Of equal merit is the stanza 
attributed to the narrator (Sn. 30). The refrain (occurring in Sn. 18-29) 
with the word deva for rain-cloud is very effective. 

Simile, metaphor, world-play, alliteration and assonance and onomate- 
pocia are employed to some degree. Both the similes used at Sn. 29 
reflect ideas quite familiar to other parts of the Canon (cp. Th\, 1184, 
Th2, 301. Ap. 60, 10 etc.). Most of the Buddha's answers to Dhaniya 
contain metaphors. E. M. Hare (Woven Cadences, pp. 218ff.) has pointed 
out word-play at Sn. 18a == 19a, 18b = 19b, 22a = 23a, and alliteration 
and assonance at Sn. 21a, 25abc, 28a, 29b and 33ab=34ab. An onomate- 
poeic effect is produced by phrases such as, sutva devassa vassato (Sn. 
30c), etc. In spite of the poetical devices employed there appear no 
signs of lateness in the language of the poem and there is much positive 
evidence to show that the poem is old. 


The metre of the poem is not uniform, but it is evident that it follows 
the pattern of the Vaitallya and Aupacchandasika — both metres of popular 
origin in which the syllabic instants are taken into account. Something 
definite can be said of only the two concluding stanzas which are in 
Vaitallya (14, 16; 14, 16). Of the remaining stanzas the sum-total of 
morae in a half-stanza ranges from 29 (Sn. 30cd) to 37 (Sn. 28ab). Some 
of the pairs of stanzas in the main dialogue agree metrically; viz. Sn. 
18, 19 contain 36, 30 morae each in their half stanzas, Sn. 20, 21; 32, 
32 and 32, 31 morae respectively. Sn. 26, 27; 31, 32 morae in each: while 
there is a disparity in varying degrees in the others viz. Sn. 22, 23 in pada 
c, only (16, 19, 16, 17; 16, 19, 13, 17 respectively), Sn. 24, 25 in padas ac. 
(14, 18, 16, 17; 13, 18, 13, 17 respectively), and Sn. 28, 29 in all padas- 
except the refrain (17, 20, 17, 17; 15, 17, 18, 17 respectively). 

In addition to Sn. 33 and 34, Sn. 32 can also be taken as a Vaitallya 
stanza with anacrusis in the pada b, as pointed out by Helmer Smith 27 . 
Sn. 20 consists of the number of syllabic instants required for the Aupac- 
chandasika metre and the rest of the stanzas contain either more or less 
morae than required for that metre. Other metrical irregularities already 
observed by scholars 28 are: anacrusis at Sn. 22b = 23b, 30b, trochaic 
pada at Sn. 24a; and 26b = 27b containing a mora too short (godharaniyo) 
for the Aupacchandasika metre. (Helmer Smith, ibid). The lack of 
uniformity in the metre of the poem supports the view that the writers 


SnA. 643. 

154 Pali Buddhist Review 2, 3 (J 977) 

of these/ ballads have been mainly guided by rhythm and not fixed metres. 
Perhaps these stanzas mark the transition from the purely popular 
versification to the more fixed Vaitaliya and Aupacchandasik metres. 
This too is in agreement with the general antiquity of the poem. 

Doctrinal Developments 

No real developments in doctrine are in evidence in the sutta. The 
ideology of the poem conforms to that of Buddhism in its earliest phase. 
The various Buddhist concepts alluded to in the Buddha's replies to 
Dhaniya belong to early Buddhism, and therefore demand no particular 
attention. The only term with a technical significance is upadhi 
(substratum of rebirth — Sn. 33, 34) which is known from early Buddhist 
times, and is met with no less than 19 times in Sn. On doctrinal evidence 
the sutta appears to be considerably old. 


External Evidence 

An interesting feature of the poem is that very few of its padas in the 
main dialogue are to be met with in other metrical works, though the 
refrain occurs at Th\. 51-54, 325. On the other hand, the ideas in the 
sutta are common to other parts of the Canon as well. The various 
metaphorical allusions (e.g. to kuti, gini, bhisi, etc.) in the Buddha's 
replies, can be compared with the numerous similes and parables in 
other parts of the Canon 29 ; e.g. the simile of the kuti at M.I., 190, Th\. 
125 etc., bhisi cp. the parable of the raft (kulla) M.I., 134, etc. The whole 
of the sixth vagga of the Eka-nipata of Th\. (Thl. 51-60) is connected 
with kuti. The occurrence of Sn. 33, 34 at S.I. has already been discussed. 
The fact that the majority of the padas in the main dialogue are not 
found repeated in other metrical works may suggest that these stanzas 
were quite distinct from the rest. Yet in contents they agree. Taking 
all the internal evidence (specially from language, syntax and metre) 
into account it may be said that these stanzas may have been anterior 
to the bulk of the Pali metrical works. 


Kasibharadvaja Sutta 

The other "Pastoral Ballad" Kasibhdradvaja Sutta is a regular akkyana 

containing narrative prose connecting the gathas. The poem itself is a 

modified parable in which the Buddha explains the Dhamma to the 

29. The Commentator gives an exhaustive list of similar occurrences (SnA. 31 11) 
which Helmer Smith has traced in the foot-notes. 

Pali Buddhist Review 2, 3 (1977) 


ploughman by employing the terminology used by him. Though the 
gathas consist of only a question and an answer to it, they fit into the 
general scheme of the narrative which is conducted entirely in prose. 
The prose which is both introductory and explanatory appears to have 
been used in order to acquaint the reader (or the listener) with the con- 
text of the verses, of which Sn. 76-80 form a separate unit (i.e. the Buddha 
as ploughman). Like all prose introductions to ballads, it is quite 
probable that at some early stage the prose of this sutta was not fixed, 
and that the singers of the ballads described the situation in their own 
words. The fact that the version of this sutta found at S.I., 172-173 
(Brahmana-Samyutta) contains these gathas verbatim, but shows a diver- 
gence in the prose 30 is in agreement with the unsettled nature of the 
prose. Further, the enhanced version of the conclusion in Sn. may be 
indicative of the relative lateness of the compilation of Sn. (as an antho- 
logy), for decidedly the shorter version in the Samyutta is the older oi 
the two. The possibility of the existence of two recensions within the 
same Theravada School is very remote, and it is quite clear that both 
versions are based on the same tradition and that the sutta in Sn. is 
merely an enlargement of the same occurring in the Samyutta. 

The formula-like phrases of the introduction and conclusion whicr. 
agree word for word with all such passages in suttas dealing with conver- 
sions made by the Buddha, and the emphasis laid on the miracle as an 
ingredient to conversion exemplifying the consequence of the iddhi- 
power of the Buddha (Sn. p. 15) rather than the teaching itself, are 
positive indications of the lateness of the prose of the sutta as contrastec 
with the gathas. There occur a few irregular verbal forms in the prose 
which are of no great value here, as they are used in all periods o! 
canonical Pali; viz. dammi, an old form used in all periods (vide Geiger 
§143) cp. Epic Sk. dadmi; dakkhinti, future, cp. sakkhinti (vide Geigei 
§152), but has the appearance of an old (Vedic) injunctive 31 ; and alattha 
S-Aor.— all occurring in Sn. pp. 15-16. One is also struck by the frequen 
occurrence of imper. 2 sg. forms in -ssu (historical) and the wealth o 
denominatives in the passage describing the miracle. 

The stanzas in the sutta are not very remarkable, and fail to reach thi 
perfection of those of the companion poem; Dhaniya Sutta. Syntactical!, 
there appears nothing worthy of comment as the verses merely compari 
in a series of simple sente nces, the counterparts in the religious life 

30. Both introduction are identical. The prose passage after the fifth stanza (Sn. 8C 
runs-' Bhunjatu bhavam Gotamo. kassako bhavam Gotamo, yarn hi Gotamo amatapphalar 
pi kasim kasatl ti. The concluding passage in S. begins with, Evam vutte Kasibharadvaj 
brahmano bhagavantam etad-avoca: Abhikkanlam bho Gotama, etc. (Sn. p. 15, 11. 17 t 
1, 1 . p. 16) and concludes: upasakam mam bhavam Gotamo dharetu ajja-t-agge pamipetai 
saranam gatan ti. 

31". Vide. Pischel §§525-6, 529, 532. 

156 Pali Buddhist Review 2, 3 (1977) 

to the various implements and actions in ploughing. On the one hand, 
the absence of exact counterparts for some items in the similes (e. g. 
Sn. 78ab), and on the other, the mention of two for the same object 
(e.g. Sn. 77bd, pahha and sati) speak further of the lack of perfection of 
the poem. 

The language of the verses is essentially gatha-?a.\i. Yet, archaisms 
are few. The only noteworthy forms—jdnemu (Sn. 76d) cp. Pkr. 
jdnimo and jdndmo, Sk. -wa/i>Pkr. -mo>P. -mu (vide Pischel §510); 
and dhuradhorayha (Sn. 79a), a peculiar double nominal form (dhura+ 
dhor-vahya s.v. P.T.S.). The sandhi in vuttir esd (Sn. 81d) is historical. 

The metre of the poem is somewhat regular; Sn. 76-80 are Anustubh 
slokas and Sn. 81-82 are Tristubh. There is anacrusis at Sn. 79a and the 
caesura is not reckoned after the seventh syllable at Sn. 82d. Metrical 
lengthening is to be seen at Sn. lie (hiri) and 8ld (sail) in addition to 
instances like puhnapekha; (Sn. 82d) for rhythm rather than metre. The 
style of the sutta has been sufficiently commented on earlier. 

The ideology of the poem does not show any late developments. The 
emphasis is on the practical aspect and the attainment of deliverance 
through moral and intellectual discipline. Qualities like saddhd, tapo, 
pannd, hiri, sati, sacca, soracca, viriya and the control over mind, body 
and speech are accepted as virtues by all early Indian schools; yet the 
training envisaged in the sutta is essentially Buddhistic. There is neither 
speculation nor metaphysics, and the teaching, however tersely expressed, 
does not show any departure from what may be expected in the earliest 

There are two terms which demand further attention; viz. yogakkhema 
(Sn. 79b) and kevalin (Sn. 82a). The technical significance of the former 
has come about by a direct semantic development from Vedic. In Vedic 
it meant "exertion and rest, acquisition and possession" (s.v. PTS), 
whereas in Classical Sk., "security, secure possession of what has been 
acquired, or insurance "as at Manu. IX, 219. According to the Peters- 
burg Sk. Dictionary, it is usually explained as "gain and support of a 
possession" and at Gaut. 28, 46 it means "property destined for religious 
purposes". But all these ideas are remote from the Pali meanings. The 
non-technical meaning of "rest from work" in Pali, is nearer Vedic, 
and the idea of "freedom from bondage" is logically connected with 
the former. The term kevalin, which occurs as many as 7 times in Sn. 
is an epithet for the "perfect one". The idea itself is very familiar to 
the Canon even in such pharses as kevalaparipunnam, etc., though the 
concept is seen best developed in Jainism (cp. kevalin, the perfect one, 

Pali Buddhist Review 2, 3 (1977) 157 

an epithet for tirthankara, and kaivalya, epithet for nirvana). The origin 
and conception of the idea is definitely pre-Buddhistic, but it need not 
be through any Jaina influence that its adoption has taken place in 

The evidence from all these sources taken singly does not yield any 
definite data as regards the date of the poem, but taking the collective 
evidence the stanzas have a claim to comparative antiquity. 


External Evidence 

It has already been observed above that the sutta is an enlarged version 
of that found at S.I., 172. The additional prose passage found on p. 15 
is seen to occur in the Sundarika Sutta at S.I., 167-168 though the 
Sundarikabharadvaja Sutta in Sn. (Ill, 4) contains no corresponding 
passage. In order to establish the relationship of these two suttas, their 
common factor Sn. 81-82 = Sn. 480-481, has to be taken into account. 
These two stanzas are repeated in three suttas in the Brahmana Samyutta; 
viz. at S.I., 167 (Aggika-), I, 168 (Sundarika-), and 1, 173 (Kasi-). This 
necessitates a comparison of these three suttas with the corresponding 
ones in Sn. 32 

In all the instances where the two stanzas Sn. 81-82 occur there is 
apparently sufficient justification for their inclusion, for the introductions 
state either that the Buddha was on his alms-round or that the brahmin 
was preparing an oblation. However, the stories contained in these 
three suttas show that at some stage or another there has set in a confu- 
sion of legend. Although it is not easy to say which sutta portrays the 
original version of the story, the influence of one on the other two is 
evident. It is significant that the section ending with Sn. 80, has very 

32. Aggika Sutta (S.I., 166-167) corresponds to, Vasala Sutta (pp. 142-151) which 
in Sn. bears the alternative title Aggikabharadvaja Sutta. The only noteworthy simil- 
arity of the two suttas lies in their respective introductions, though they differ widely 
in details. The nidana (scene of the sutta) in Sn. is Savatthi whereas it is Rajagaha in 
S. In both suttas the Brahmin is said to be preparing for a fire-sacrifice though it is stated 
in different words (Sn. p. 21, aggipajjalito hoti, ahuti paggahita; S.I., 166, sappina payaso 
sannihito hoti, 'aggiijtjuhissami,aggihuttam paricarissami'ti). In Sn. the Brahmin abuses 
the Buddha whereas in S. he offers the payasa (in verse) which the Buddha refuses with 
the stanzas Gathabhigitam, etc. The stanzas in the two poems are entirely different 
though the phrase, najacca hoti brahmano (S.I., 166 32 , Sn. 136b, 142b) is common to 
both. The concluding prose is the same. 

The introductions to both versions of the sutta recording the conversation between 
the Buddha and Sundarikabharadvaja are almost identical but only 5 of the 10 stanzas 
in S. have parallels in the 32 stanzas in Sn.; viz.; S.v. 1 =Sn.462, (v. 8), S. v.2//Sn 463 
(v. 2ab=Sn. 463ab), S. v. 3//Sn. 459, 479 (v. 3bc=Sn. 459bc), S. v. 4-5 =Sn. 480. 481. 
The miraculous incident (//Sn. p. 15) is related at S.I., 168-169, following which occurs 
a set of 5 stanzas, both of which being absent in Sn. The conclusion is identical in both 
works. (Sufficient has been said of the Kasibharadvaja Sutta, earlier). 


Pali Buddhist Review 2, 3 (1977) 

little connection with the rest of the sutta, in the Kasibharadvaja Sutla. 
The discourse ends there, and the line, 

etam kasim kasitvdna, sabbadukkha pamunccati, 
affords a fitting climax. Unlike the Sundarikabharadvaja Sutta (both in 
S. and Sn.) the story is not centred on ' 'what is left of the sacrificial 
cake" (havyasesa), which has a greater mysterious significance than 
payasa (Kasi-), a thing mentioned only at this secondary stage. The 
difference in metre of Sn. 81-82 from that in the previous stanzas, and 
the uniformity in this respect, of all the stanzas (including these two) 
of the Samyutta version of the Sundarika Sutta (and to some extent 
the poem in Sn.) tend to emphasise the fact that Sn. 81-82 didnot originally 
belong here? 1 

In view of all this evidence, both internal and externa], it may be 
concluded that (a), the sutta consists of two different elements (i) Sn. 
76-80 with the prose introduction which at some stage was not in any 
fixed form, (ii) the two stanzas Sn. 81-82 and the prose on pp. 15-16 
which constitute a subsequent addition, (b), the sutta has been greatly 
influenced by another sutta (probably the Sundarika Sutta of S.), the 
main theme of which was the exemplification of the miraculous powers 
of the Buddha, (c), the sutta in its present form has been included in 
Sn. at a d&te much later than that of the composition of the stanzas, 
and (d), the sutta probably is later than the Brahmana Samyutta, if it 
has been influenced by the Sundarika Sutta. 

33. The Sundarika Sutta (S.) presents a more coherent narrative of the incident 
while the parallel version in Sn. is either an amalgamation of two suttas, one of which 'was 
based on the first 5 stanzas of the sutta in S. or an altogether different sutta which has 
incorporated a greater part of the legend i.e. without the miracle, as well the first five 


John D. Ireland 

In the Pali Buddhist Review I p. 2 there is a translation of the Buddha 's- 
brief instruction to Bahiya (Ud. I 10) by Bhikkhu Nanavlra. As this 
difficult passage is of considerable interest it was thought worthwhile to 
attempt another and fuller translation of it. In the PTS edition of the 
Udana the Pali text is hopelessly corrupt and this translation is based 
upon Woodward's reconstruction of it from the parallel passage in the 
Samyutta Nikaya (S. IV p. 73) and the Udana Commentary (p. 92). 
The italicised portion is the amplification of the text from the 

In addition there is also translated the verse-udana, the inspired 
utterance the Buddha gives out at the end of the Bahiya Sutta, followed 
by the udana of Ud. VIII. I, which is evidently a prose gloss on the 
former. Taken together the,y provide an illuminating description of the 
nature of Nibbana, bearing upon the Buddha's brief instruction to 

It might be of interest to study in conjunction with these the Buddha's 
advice to Malunkyaputta in the Samyutta (S. IV p. 72f) and also the 
Millapariyaya Sutta (M. 1). In the Samyutta the passage is not intro- 
duced so abruptly, but the Buddha leads up to it by questioning 
Malunkyaputta which clarifies what is intended. And then the verses 
expand on Malunkyaputta 's understanding of it. 

Basically the message is to bring an end to the deep-rooted craving, 
attachment and fondness for those things cognised by the senses, 
experienced now, remembered and yet to be experienced in the future, 
at this very moment now. And when this is realised one does not identify 
oneself with the pleasure and delight involved, even in the most refined 
and subtle achievements of meditation experience. They are not thought 
of as identical with or a possession of the 'self; as 'me' or 'mine'. 
Udana p. 8 restored: 

Tasmat iha Bahiya evam sikkhitabbam: 
ditthe ditthamattam bhavissati, sute sutamattam bhavissati, 
mute mutamattam bhavissati, vinnate vinnatamattani bhavissatl'ti. 
Evan hi te Bahiya sikkhitabbam. 

to cut my own artery; 

thorough application of mind arose in me 

the danger was revealed and then 

weary with the world was evenness established. 

Then my mind was Free! 

See the Dhamma's normality! 

Possessed is the triple knowledge, 

done is the Buddha's Sasana. (405-410) 
Sappadasa tried to kill himself out of despair because of his wandering 
mind. He was born as the son of King Suddhodana's ceremonial priest 
and therefore of brahmin stock. When the Buddha returned to his own 
people to teach them, he obtained confidence and went forth. He was 
overpowered by defilements of mind and so could not win one-pointed- 
ness of mind. Finally, he became so distressed that he got to the point 
of committing suicide but then insight arose and Arahantship was attained. 
Declaring his perfect knowledge, he uttered the above verses. 


N. A. Jayawickrama 


The Pabbajja and Padhana Suttas and the Vatthu-gatha of Ndlaka Sutta 
and Parayana Vagga and part of the epilogue to the latter can be classed 
as narrative pieces in Sn. In addition to these there are other isolated 
narrative verses (Sn. 30, 251-252, etc.), which the Commentator himself 
attributes to the sahgltikara. Out of these narrative pieces, the Vatthu- 
gatha of the Parayana have been fully discussed in the general remarks 
on that vagga. An attempt will now be made at a more detailed examin- 
ation of the Pabbajja, Padhana and Nalaka Suttas. A brief reference 
has already been made to them, and Winternitz's significant statement that 
they form the earliest beginnings of a life of Buddha in verse, has been 
noted. A little more has been said about these suttas in the general 
discussion on the Mahavagga, and the arrangement of the suttas in it. 


Pabbajja Sutta 

The Pabbajja Sutta 1 is essentially a narrative ballad, which on account 
of the highly interesting dialogue it contains can be called a dialogue- 
ballad at the same time. The whole sutta is built upon the event of Bim- 
bisara's first meeting with the Buddha. The first three stanzas serve as 
an introduction to the narrative, which proceeds throughout in the 3rd 
person, and the other 17 (Sn. 408-424) constitute the body of the ballad. 
It will be noticed (later) that these introductory verses did not form an 
integral part of the poem. The dialogue-stanzas of the sutta are of a 
highly dramatic character. The narrative-stanzas interspersed with the 
dialogue, describe in successive stages the events leading up to the point 
when the respective characters represented in the sutta make their state- 
ments. It is not improbable that this poem was a regular dramatic ballad, 
in which the narrator recited the narrative stanzas while others sang the 
respective stanzas assigned to the various characters; for, in many respects 
the narrative verses closely resemble the prose narrative element in the 
regular "A^yana-ballads" of Sn. (e.g. Kasibhdradvaja Sutta); and the 
dialogue stanzas, the dialogue element in such suttas. The description of 
the chang e of scene and events enables the listener to follow the dialogue 

I. The Commentary {SnA. 381) ascribes the sutta to Ananda. 

4 Pali Buddhist Review 3, 1 (1978) 

From the analogy of the "Akhyana-type" of mixed ballads it may be 
argued that only the dialogue (Sn. 410-411, 416, 420-424) formed the 
original ballad and that the narrative stanzas were merely a versification 
of earlier extant prose, which was perhaps like the narrative prose of some 
of the suttas in Sn. However, this cannot be established with any degree 
of certainty. Neither can it be said whether the poetical forms in the 
narrative stanzas are artificial forms based on the prose or not. Yet, in 
two instances (Sn. 406, 417-419) the narrative verses closely resemble the 
standard form of expression in prose canonical Pali. 2 This coincidence 
is not a mere accident. 

As regards the forms themselves in these narrative-stanzas they present 
no divergence from the normal gathd-Pali. Linguistic data suggest an 
earlier date for the dialogue-stanzas. The uniformity of metre (Anustubh 
Slokd) is perhaps due to the attempt on the part of the writer of the 
akhyana in verse to present a uniform ballad. The striking forms in the 
dialogue-stanzas are:— Sn. 410, bhonto, braha the use of which is entirely 
restricted to poetry (s.v. PTS.), pekkhati; Sn. 411, the sandhi, mcakula- 
m-iva, which though inorganic is essentially old Pali; Sn. All, anlka of 
direct Vedic origin bhunjassu, akkhahi; Sn. 422, the adverbial usage of 
ujum which is archaic (v.l. uju. cp. Mvastu. nija- which Neumann calls a 
misunderstanding of the old Pali) Sn. 423, kame abhipatthaycm (a) abhi 
construed with ace. (b) the old p. pr. in -am, Sn. 424, kamesu. . an historical 
construction belonging to old Pali and datthu, irregular archaic absolutive. 

There exists no early prose record of this incident. According to later 
tradition (SnA. 382 ff., J.I, 66 and DhA. I. 85) the meeting between Seniya 
Bimbisara and Gotama took place prior to the Enlightenment. The 
reference made to him as the Buddha (Sn. 408a) and cakkhuma (Sn. 405 b ) 
need not imply any contradiction, for even prior to the Enlightenment 
Buddha may be spoken of in such terms by later writers. Yet, the refer- 
ence here is to the personal Buddha. 3 As a rule, the term as referring to 
the personal the Buddha was not very popular in the earliest portions of 
the Canon, where, invariably, he is called Bhagava or Tathagata. But 
its use as, "the Enlightened" or "the Awakened" is early, e.g. S.I. 35, 
60, A. IV. 449, Sn. 622, 643, 646, etc. Besides this the occurrence 

\ H*$- 406 cp " MJ > 179 > S ' V - 350 > A - n ' 208 > etc - sambadho gharavaso raja- 
patho, abbhokaso pabbajja. 

_(b) Sn. 4P-419 cp. D.I, 50 II. 73, A.V. 65, Vin. 1, 231, 242, etc bhaddSni bhaddani 

yanam yojapelva, Bhaddani yanam abhiruhitva, bhaddehi bhaddehi yanehi niyyasi, 

yena . . tenapayasi, yavalika yanassa bhumiyanena gantva, yana paccorohitva, pattiko 'va 

yena bhagava ten 'upasankami upasaAkamitva bhagavata saddhirn sammodi, sammodamyam 

katham saranlyam vitisaretva ekamantatn nisidi. 

3. The term Buddha occurs 39 times in the gathas of Sn. . . As many as 25 refer to 

the Persona) Buddha, and the other 14 to Buddha in the impersonal sense. The term 

Sambuddha is met with 19 times. 

Pali Buddhist Review 3, 1 (1978) 5 

of the phrases, akinnavaralakkhano (Sn. 408 d ) and lakkhanasampannam 
(Sn. 409°) may be accepted as indicative of a certain amount of develop- 
ment in the concept of the Buddha. The nominal forms, pasadasmim 
(Sn. 409 b ) and rajino* (Sn. 41 5 d ) apparently belong to a considerably late 
stratum of Pali. All this evidence does not go to prove the lateness of 
the entire poem, but that probably the narrative verses may not be as old 
as the rest of the poem. This may still be maintained in spite of the 
general uniformity of the poem in many respects. It is clear that the event 
reported in this sutta took place prior to the Buddha's Enlightenment 
(vide Sn. 424°) and that it was the first time that Bimbisara met the Buddha. 
This is further attested to by the strong tradition preserved in the Nidana- 
katha of the Jaiaka. 

However, there are a few discrepancies in the poem. The occurrence 
of the term Buddha at Sn. 408 has already been discussed. It is to be 
noted that the version of this sutta in the Mahavastu (Mvastu. II. 198 ff.) 
does not refer to him as the Buddha. Again according to the Pabbajja 
Sutta, the Buddha had a following even at this stage — ndgasangha purak- 
khato, Sn. All. Neither Mvastu. nor the condensed version in the Nidana- 
katha makes any mention of a following or a sarigha. In fact Sn. 420- 
421 are represented by only one stanza in Mvastu., 

Udagro tvam asi rajnah asvaroho'va selakoj 
dadami bhogam bhunjahi, jatlm cakhyahi prcchitojj 

It may be quite possible that in this instance Mvastu. preserves an older 
tradition while the two stanzas in Sn. indicate an expansion on a different 
line. This is further borne out by the strange resemblance of Sn. 420 ab 
to the oft-recurring prose formula, daharo hoti, yuva susu kdlakeso 5 
bhadrena yobbanena samannagato . . . .M. I, 82, D, I. 115, A, II, 22, III, 
66, etc. Some of these apparent contradictions may be ascribed either 
to later accretions or to a confusion of the tradition at some early stage. 
The latter possibility is more plausible when all the other available evidence 
is taken into consideration. 

Though both versions narrate the same event, the Pali and BSk. show 
definite signs of independent development from their original source, if 
such a version did exist. In the case of Pali this has been effected mainly 
by the association of the forms of expression and formulae pertaining 
to the standard prose idiom. Many of the discrepancies in evidence in 
Sn. can be thus explained on this basis. Two such instances have been 

4. The only other occurrence of rajino in Sn. is at 209. in the Brahmanadhammika 
Sutta. To say nothing of that sutta, but taking Sn. 299 independently, its comparative 
lateness is evident from the late word vipallasa and the (late?) artificial nominal form 
viyakararn occurring in it. 

5. Also susukalakeso which is explained by Commentaries as, 'with very black hair*. 

6 Pali Buddhist Review 3, 1 (1978) 

noticed earlier (above). The three introductory stanzas are replaced 
by a brief prose sentence in Mvastu. which states that the Bodhisattva 
leaves Arada Kalama and repairs to Rajagrha. There is nothing corres- 
ponding to Sn. 413 in Mvastu. The number of instances in which the 
pa das of this stanza are seen to occur in other Pali metrical works (Hare, 
p. 195) perhaps indicates a possible explanation for the presence of this 
stanza here. The stanza Sn. 416 is expanded into two verses in Mvastu. 
and Sn. 417 into three. The stanza corresponding to Sn. 418 in Mvastu. 
is totally different from the Pali which bears kinship with the prose for- 
mulae. Again, Sn. 424 ab (cp. Th\. 458, Thl. 226, etc.) has no parallel in 
BSk. Though the dialogue between Bimbisara and Buddha (Sn. 420-424) 
is found in a more condensed form in Mvastu., it does not end where it 
stops in Sn., but continues with two more stanzas in which Bimbisara 
solicits Buddha's promise to visit him after the Enlightenment. 

The story in the Nidanakatha (J. I, 66) is not very helpful in the analysis 
of this sutta, as it is even posterior to SnA. which it mentions. 6 

Other internal evidence consists of an examination of the places ment- 
ioned in the sutta. Rajagaha was connected with Buddha's early career, 
and was one of the earliest centres of Buddhism. The peak Pandava 
was situated in the line of hills which formed a natural fortification to 
the city, giving it the name Giribbaja (see also DPPN). The Sakiyas 
are spoken of as a family of the Aditya clan inhabiting the Himalayan 
sector of Kosala. Legend has not yet grown round them making them 
an all powerful clan. They are merely a kula in Kosala. This too 
supports the general antiquity of the poem. However, the evidence at 
hand shows that the dialogue-stanzas preserve an older stratum than the 
narrative verses which betray signs of further development. It is quite 
probable that the three introductory stanzas which cannot be traced in 
Mvastu. were still later than the narrative verses. On account of the 
general consistency of the poem in language, metre, style and syntax 
it is not possible to say by what length of time these stanzas were separated; 
yet it must be agreed with Winternitz that this sutta is a precious remnant 
of the ancient ballad-poetry from which the epic of the life of Buddha 

6. A comparison of the two is interesting merely from the point of view of tradi- 
tion. In the Nidanakatha the dittos see the Buddha and inform the King, and it is they 
who speculate whether he is a deva, human being etc. . It is described how the Buddha 
loathed the meal he obtained by begging alms, and he finally ate it after self-admonition. 
Bimbisara visits the Buddha and is impressed by his bearing — iriyapathasmini pasiditva 
— and offers him all comforts which the Buddha refuses. He finally solicits a promise 
from the Buddha to visit Magadha after the Enlightenment. Buddha then goes to 
Alara Kalama, and Uddaka Ramaputta and finally practises austerities — mahapadha- 
narji padahitukamo mahfipadhanaip padhanesi. Subsequent events are next recorded 
in the NidanakathS. 

Pali Buddhist Review 3, 1 (1978) 

Padhana Sutta 

The Padhana Sutta appears to be foreshadowed in the last stanza of the 
Pabbajjd Sutta-padhanaya gamissami, (Sn. 424 c ). The two sulfas are 
closely connected with each other, but in spite of Sn. 424 c it is doubtful 
whether they aim at a connected narrative, though they ostensibly appear 
as such. In view of the changes that the Pabbajjd Sutta has undergone 
at editorial hands it may be surmised, though it cannot be established with 
certainty, that it served as an introductory sutta to Padhana in Sn. and 
that Sn. 424 c was a mere coincidence. Mvastu. hardly throws any light 
on this, on account of the fact that the sutra there, while preserving some 
of the primitive characteristics, also shows an expansion on a line different 
from that of Sn., and besides, some of the sections that are placed between 
the two suttas contain much irrelevant matter (such as jatakas). How- 
ever, it is significant that the next sutta in Mvastu. deals with incidents 
following Buddha's departure to Uruvilva leaving Udraka Ramaputra. 
The pada, padhanaya gamissami is common to both Pali and BSk. (prdha- 
naya gamisyami— Mvastu. II, 199 18 ), and must necessarily be old, but it 
seems to have been partially responsible for the prefacing as it were, of the 
Padhana Sutta with the Pabbajja— besides taking into account the logical 
sequence of these two surviving ballads. 

These two suttas represent but two of the major episodes in the eventful 
period of Buddha's early career, the one, at best being a record of Bimbi- 
sara's first meeting with the Buddha, and hence be more appropriately 
termed "Bimbisarapratyudgama" (vide PBR 1, 3), while the other an 
allegorical representation in ballad-form, Buddha's conquest of evil. 
On the strength of the evidence from these two suttas alone, the relevant 
intervening incidents recorded in Mvastu., SnA., DhA., and Nidanakatha 
(J. I, 66), however late some of these accounts may be, cannot be all 
brushed aside as subsequent accretions in the course of development of 
the story of the Buddha. Although it is quite obvious that the later 
accounts are highly embellished versions of the life of the Buddha, the 
fact that only these two important events of the renunciation and the 
quest of peace by asceticism are preserved in the form of ballads, neither 
precludes the possibility of the early existence of more ballads of this 
nature, nor establishes that these suttas contain a complete record of 
Buddha's early sojourn as an ascetic. Yet, "the wholesome austerity" 
as pointed out by Chalmers (p. xix) evinced in these suttas gives them 
precedence over all other extant accounts. 

8 Pali Buddhist Review 3, 1 (1978) 

The Padhana Sutta, like its companion poem, the Pabbajja Sutta, is a 
narrative incorporating dialogue-stanzas. The occurrence of the 1st 
person in the opening stanza (Sn. 425*— mam) cannot be reconciled with 
the 3rd person in the narrative at Sn. 429 cd , 

ima gdtha bhanam Maro atthd Buddhassa santike. 

It has been correctly pointed out by Katre that mam is an error for nam, 
which has its antecedent at Sn. 408 a . 7 The fact that this line has no exact 
parallel in Mvastu. makes the verification of Katre's suggestion rather 
difficult; and furthermore, the uniformity of the Pali Mss. which read 
mam throughout shows that the "error" has set in at a very early stage. 
This confusion is also noticeable in Mvastu. though the exact parallels 
are not found there. (The account at Lai. 299 ff. is of no value as it offers 
no parallel to Sn. or any other Pali version). The opening stanza in 
Mvastu. reads, prahanam prahitam mayd (II, 238 4 ), but the narrative 
reverts to the 3rd person in the sixth stanza, 

imam vacam bhane Maro, Bodhisattvasya santike. 

This coincidence, besides establishing for certain the common origin of 
the two versions throws some light on the narrative element in this sutta. 
A glance at the sutta shows that it contains comparatively few narrative 
stanzas, (viz. Sn. 425, 426 ab , 429 cd 430 i,b and 449) as contrasted with 
Pabbajja Sutta. The dialogue-stanzas at Sn. 425 ab -429 ab can be taken as 
forming three complete stanzas; thus; 

1. Kiso tvam asi dubbanno, santike maranam tava, 
sahassabhago maranassa, ekamso tava jlvitam. 

2. Jiva bho, jlvitam seyyo,jlvam punnani kahasi, 
carato ca te brahmacariyam aggihuttam ca juhato. 

3. Pahutam ciyate puhnam, kim padhdnena kahasi. 
Duggo maggo padhanayu dukkaro durabhisambhavo. 

Similarly Sn. 430^ and 431 can conveniently form a stanza of six pddas 
like Sn. 434. (It is not possible to arrange Sn. 430 cd -434 e£ into five stanzas 
without breaking up complete sentences and disturbing the harmony of 
the poem). It is quite probable that at some stage the sutta consisted of 
only the dialogue, the narrative stanzas being a versification of older 
prose. The presence of narrative verses in Mvastu. makes it quite clear 
that this has taken place at a very early stage. The confusion in the 
narrative may be ascribed to that same period. 

7. Neumann (Reden, p. 469) equates tarn main to tarn' mam (=tam Imcm — ana- 
phoric, like so 'ham) which is a brilliant suggestion which explains the whole discre- 
pancy, though the exact idiom is not to be met with elsewhere. 

Pali Buddhist Review 5, l ^iv/o) 

The opening dialogue-stanzas quoted above are the words of Mara, 
and Buddha's reply commences at Sn. 430 cd and ends at Sn. 440. The 
next five stanzas are in the form of a soliloquy, and the end of Sn. 443 
marks the complete defeat of Mara, while Sn. 444-445 constitute the 
"victorious resolution of the hero" (Katre). The next three stanzas 
representing Mara's acknowledgment of defeat appear to be a subsequent 
addition. They are not found in Mvastu.; but it is stated at S.I, 122 that 
Mara was on the Buddha's trail for seven years waiting for an opportunity 
to seizehim, but with no success— olarapekkho, olaram alabhamano. Later, 
in the same section (S. I, 124), he acknowledges defeat and utters the 
identical stanzas at Sn. 447-448. It is quite probable that Sn. 446 is a 
versification of a passage corresponding to that at S. I, 122 while the next 
two stanzas were perhaps taken from the same source as S. The final 
stanza of the poem {Sn. 449) roughly corresponds to that in Mvastu, and 
forms the narrator's conclusion. 

There is no doubt that the sutta is old, but the whole of it cannot be 
assigned the same antiquity. Some austerities practised by the Buddha are 
mentioned at M. I, 242 ff. Here the Buddha relates how he gradually 
gave up self-mortification and fasting. He took food in gradual quantities 
till he became strong again. The pancavaggiyas left him saying, bahuhko 
samano Gotama, padhanavibbhanto avatto bahullaya. ("The ascetic 
Gotama has swerved from his austerities and has reverted to a life of 
luxury"-M. I, 247, cp. M. I, 17-24, 114-118, 167, etc). Then he evolved 
the four jhanas and realised the three vijjas. There is no mention of 
Mara in the Majjhima account. S. 1, 103 speaks of Mara as having visited 
him when he was seated at the foot of the Ajapala banyan tree, after 
attaining Enlightenment. Lai. devotes a whole section (Maradharsana- 
parivartah—Lal. 299-343) to the Buddha's conquest of Mara. (S. I, 124 
will be discussed later). Besides these there are numerous late accounts 
which deal with this topic in great detail, e.g. J. I, 71 ff. BvA. 239, SnA. 
391, DhA. II 195, etc. Actual battles are spoken of between the legions 
of Mara and the Buddha, and many of the late accounts make paramitas 
combat Mara. 

The only version which bears a close resemblance to the Padhana Sutta 
is the section at Mvastu. II, 237 ff. Their common origin has already 
been hinted at. Amo ig the numerous incidents reported in Mvastu 
between the two sutu.o ... re, ponding to Pabbajja and Padhana, there 
occurs a description of severe austerities practised by the Buddha (II 
231 ff.). It is stated that he lived on one kola (Pali, kalayal) a tila anc 
a tandula each a day for three successive periods of 18 months each, anc 
no food at all for a further period of 18 months making up a total of so 


Pali Buddhist Review 3, 1 (1978) 

years which agrees with all other accounts (cp. Nidanakatha). An old 
parallel to this is found at M. I, 245; thokam thokam aharam aharesim, 
pasatam pasatam yadi va muggayusam, yadi va kulatthayusam-pe- . 

The prose introduction to the Prahana Sutra occurs at Mvastu. II, 237, 
in which it is stated that the Evil One visited the Bodhisattva while he was 
practising; austerities (duskaracarikam carantam) at Uruvilva on the 
banks of the Nairanjana. Comparing this narrative with M. I, 245 the 
striking similarity in the essentials, leaving aside the extraneous matter 
in the former, is the common basis of their origin that they point to. 
The main purpose of the narrative in Mvastu. being the linking up of 
various jdtakas it is very unlikely that much attention was paid to the 
narration of the present story. This being the case exaggeration and 
poetic embellishment find no place in this part of Mvastu. 

It is significant that it contains no passages corresponding to Sn. 427 rb , 
438 cd , 440, 44i cd , 442, 443 ab , 444 c , 445 cb , and 446-448. Of these Sn. 
427 ab is a descriptive line emphasising the odds against Gotama, while 
438 cd is a phrase found in a slightly different form in prose, attanam 
ukkamseti pare vambheti, M.I. 402, A. II, 27, etc. and is probably an 
importation to the sutta. This is further strengthened by the phrase 
Jabho siloko (Sn. 438 a ) which closely resembles the familiar 
phrase Idbhasakkarasilokanisamsa, whereas Mvastu. reads lobha for 
labha (probably a scribe's error). 

The absence of the stanza Sn. 440 in Mvastu, its rhetorical effect lending 
a realistic touch, and the occurrence of line cd. at Thl, 194, J. VI, 495 
make it appear rather suspicious in the eyes of the reader. Scholars 
have discussed at great length the phrase, esa munjam parihare. ("Look 
you, I bear the munja grass" — Neumann, esa—Du da, Hare, "See, 
I bear munja grass"). Otto Schrader (JRAS. 1930, pp. 107-109) refers 
to Pischel's misinterpretation of the phrase as Ich verschmahe das Schilfrohr 
("I refuse to take the reed"); so does Oldenberg reject it (ZDMG 1908, 
p. 594). He quotes five passages from Gobhila Grhya Sutra, Katyayana 
Srauta Sutra and Satapatha Brahmana to show the connection oipariharati 
with muhja-mekhala or muhjayoktra — i.e. wearing a girdle. Dr. Schrader 
disagrees with Oldenberg 's view that Sn. 440-442 is a soliloquy interrupting 
the direct speech of the Boddhisattva to Mara and says that these verses 
are calculated to frighten Mara though Sn. 442 may not seem to be directly 
addressed to him. Basing his argument on Sn. 431 d he says Maram 
here is a poetical substitute for M&ra tvam and interprets the phrase as 
"I take this vow (to conquer or to die, caring nothing for life"); cp SnA. 
39. K. Chattopadhyaya (JRAS 1930, pp. 897-898) agrees with Schrader 
but prefers to translate it as "I gird up my loins" (which meant that he 

Pali Buddhist Review 3, 1 (1978) 


would use his utmost vigour in his spiritual fight). He equates the 
passage to Eso' ham parikaram badhnami (Venisamhara IV). 

The section Sn 439-444 is represented by only four lines at Mvastu. II, 
240 and bears definite signs of enlargement. Judging from the fact that 
it was not customary for Mvastu, to summarise and condense, and that 
it often contains expansions of passages found in brief in Pali it cannot 
be said that Mvastu. here contains a summary. The absence of a stanza 
corresponding to Sn. 442 in BSk. and the fact that it consistently refers 
to a real army and not an allegorical representation as in Sn 436-438, are 
probable indications of the lateness of this stanza. Sn. 445 -appears « 
a familiar expression adapted from the prose. The section Sn. 446-448 
has already been dealt with (above). Thus, many of the lines in Sn 
which have nothing corresponding to them in BSk. appear to be poetical 
flashes for embellishment which perhaps did not belong to the earliest 
form of this sutta on which were based the two versions in Pah and BSk. 
The Mvastu. too shows an expansion, which however, as in the case of 
the Pabbajja Sutta has proceeded in a different direction from that in Sn. 
Besides numerous other padas and parts of stanzas which have no counter- 
part in Sn. the stanza immediately preceding the concluding verse does 
not occur in Sn., but can be traced in Dh. 26 ab and Thl. 883. An instance 
of a divergence in simile is seen in amapatram va ambuna which is meant 
to correspond to amain pattam va amhand. (Sn. 443 ). Again, Sn. 446- 
448 need further investigation. If Sn. has borrowed the last two stanzas 
of this section from S. it follows that this part of the Padhana Sutta is 
later than the Mara Samyutta. Taking into account the propensities 
of Buddhist writers to incorporate gathas wherever possible, it would seem 
natural that S. should also contain Sn. 446 in verse. From this it may 
be deduced that Sn. 446 was not known in gatha-fotm by the tune of the 
compilation of the Mara Samyutta. Hence any inference that Sn. has 
directly borrowed them from S. would be erroneous. 

On the other hand, from the aspect of the development of the Mara- 
legend S. I, 124 appears later than *. Here three of Mara's senas in Sn. 
tanha, arati (cp. arati BSk.) and kama (viz. fourth, second and first) are 
personified as his three daughters Tanha, Arat. and Raga who 
attempt to allure the Buddha. Thus, on the whole the Mara Samyutta 
appears to be later than Sn. 

Judging from the fact that Sn. 446-448 are not known to Mvastu. it 
may be inferred that at some stage the concluding stanzaSn.449 occurred 
immediately after Sn. 445 and that with the introduction of the new stanzas 
the concluding narrative verse was shifted to occupy its present position 
The pOda b, vlna kaccha abhassatha ("the lute fell from his arm-pit ) 


Pali Buddhist Review 3, 1 {1978) 

suggests a confusion of legend, the origin of which seems obscure. The 
Commentary (SnA. 393-394) states that it was this vlna (called Beluva- 
pandu) that Sakka presented with to Pancasikha. Yet, this does not 
solve the question of how Mara came by a vlna. The parallel pada in 
Mvastu. which reads, vinasam gacchi ucchrit i ("his pride was all shattered ') 
probably expresses the original idea that may have existed, prior to Ihe 
importation of the vlna from the developed legend which speaks r his 
daughters as playing instrumental music as a part of their wiles. This 
phrase perhaps dating not earlier than the time of the incorporation of 
Sn. 446-448, a confusion as it may seem, is at best a master touch of poetic 
fancy bringing the suite to a dramatic climax. 

An examination of the internal evidence from language and syntax, 
metre and ideology confirms what has been already noticed. The idiom 
throughout is old gdthd-Pdli, and from the point of syntax the following 
expressions depict a very old idiom: — Nadim Neranjaram pati-Sn. 425 b , . . 
seyyo . .yan ce..-Sn 440 cd , ma main thana acdvayi-Sn. 442 d ; etc. The 
sutta is full of archaic nominal and verbal forms e.g. Namuci (for Mara) 
-Sn. 426 a , 439% which is old Vedic (Neumann, p. 469), amhana-Sn. 
443 d , kahasi-Sn. All*, 428 d (<karsya- Geiger, 54.4, 153.1), socare -Sn. 
445 d , nddhigacchissam -Sn. 446 c (is- Aor.), attha -Sn. 429 d , anupariyaga 
-Sn. 447 b (\/Aor.), bhanam -Sn. 429°, and vinayam. -Sn. 444 d , etc. 

The metre throughout is old Anustubh Sloka. The few metrical 
irregularities are: — anacrusis at Sn. 428 a , 43 l a , even quarters at Sn. 
435 a , 440\ 443\ 428 a , 439 c and 444 c . 

No developments in doctrine are noticeable. The thought and ideas 
embodied in the sutta are distinctively old. Confidence (saddha) and 
viriya and pannd -Sn. 432 have no special technical significance which is 
to be seen even in very early works. Other qualities mentioned are, 
cittappasdda, sati and samadhi - Sn. 434. The severe austerities referred 
to at Sn. 433-434 are characteristic of the times. Another important 
concept is yogakkhema (already discussed) which has been seen to pertain 
to the earliest phase of Buddhism. Mara is called yakkha at Sn. 449. 

All this evidence shows that the sutta is old as a whole; but as observed 
earlier, Sn. 446-448 should be considered as being later than the rest of the 
poem. It is also probable, from the analogy of the Pabbajja Sutta, that 
the narrative element in verse need not have formed an integral part of 
the poem and that the nucleus of the sutta was the dialogue. 

Pali Buddhist Review 3, 1 (1978) 


Nalaka Sutta 

The Nalaka Sutta consists of two parts, the introductory vatthu-gatha 
(Sn. 679-698) and the dialogue-discourse (Sn. 699-723) dealing with 
moneyya— the state of a muni. As the vatthu-gatha present a multiplicity 
of problems they call for separate attention. Generally, introductions 
to old Pali ballads are considerably later than the poems themselves 
(vide PBR 1, 2, p. 86 and vide E. J. Thomas, Life of Buddha, p. 38). It 
will be seen that this is clearly borne out by the vatthu-gatha in spite of the 
fact that they are in verse (also cp. vatthu-gatha of Pdrayana). Unlike 
the introductions to many other suttas which narrate the incidents leading 
up to their preaching, these gathas have little bearing on the sutta proper. 
There is a difference in point of time in the sequence of events in the two 
parts of the sutta. As regards characters in the v.g, a close parallel is 
offered by the Par ay ana Vagga, for Nalaka plays the same role as Pihgiya 
in Par (vatthu-gatha, puccha and epilogue) while Asita's position here is 
very similar to that of Bavarl. 

Despite the slender connection between the two parts, the internal and 
external evidence establishes beyond doubt that a fusion of two independent 
ballads has taken place, as in the case of the Sela Sutta (Sn. pp. 102 ff. 8 ) 
and that the two components were separated in point of time. 9 

The language, style and metre of the vatthu-gatha differ considerably 
from those of the sutta proper. At the same time there is a marked 
tendency towards the g rowth of a developed Buddha-legend, which is 
totally absent in the discourse. This is evident from the reference to the 
thirty-two marks (vide E. J. Thomas, ibid.) and the occurrence of the 
term Bodhisatta at Sn. 683. 10 The general tone of this part of the sutta 
with its description of the devas rejoicing at the birth of the Buddha and 
Asita's prophecy is that of a later piece. 

Language and style.— There are many late and Sanskritic forms lying 
side by side with equally numerous very old forms: e.g. cittimkaritvd, 
atirivc kalyarupo (Sn. 680), lomahamsano, maru (Sn. 681), manussaloke, 
hitasukhataya (Sn. 683), the epithets in Sn. 684, avamsari (an analogical 
form 'Sn., 685), sik ht— fife— , the simile in the lines be (Sn. 687), patiggahe 

8. Vide Katre. He states that the intervening prose between Sn. 567 and 568 
is ''solely due to the fusion " of two different ballads. 

9 Vide E. J. Thomas, ibid, p. 39 "The question of the sutta is quite different trom 
the question of the origin of the legend (Asita's) and its becoming attached to this 

10.' Though "the Bodhisattva doctrine probably originated in the second century 
bc " (Har Dayal— The Bodhisattva Doctrine in Buddhist Sanskrit Literature, p. 43) 
the term is older; but it does not reflect the oldest stratum of thought in Pali Buddhism. 


Pali Buddhist Review 3, 1 {1978) 

(an artificial form -Sn. 689), Sakyapuhgavam (Sn. 690), gamanam (= 
maranam), akalyarupo (Sn. 691), the sandhis, isi-m-avoca, capi-m-assa, 
the phrase adhimanasa bhavatha (Sn. 692), kalakiriya, asamadhura (Sn. 
694), and hilamanasena (Sn. 697). While the late forms suggest a date 
for the vatthu-gatha the old and archaic forms handed down from an 
older period as the standard vehicle of poetic expression require no 
comment. The sutta itself (iVa/a/ca-discourse) is marked by a total 
absence of late forms. 

The style of the discourse is quite different from that of the vatthu- 
gatha. The miraculous and the semi-supernatural element is a dominant 
feature in the latter. The narrative in addition to its highly ornate 
character is extravagantly descriptive and abounds in simile and metaphor, 
e.g. Sn. 686-687, etc. There are also instances of the same statement 
being repeated in similar words, e.g . Sn. 687, 689, which have the appear- 
ance of commentarial gloss. The sutta proper is written in a much 
simpler style. 

Metre. The Sutta proper (Sn. 699-723) is in uniform Sloka metre like 
the Pabbajja and Padhdna Suttas while the vatthu-gatha are in a jumble of 
metres i.e. Sn. 681. 682, 684, 688-690 (except 688 b ) are in Tristubh with 
jagatipadas, and Sn. 679-680, 683, 685-687, 691-698 are in a metre of their 
own with the Tristubh rhythm continued. Though it is generally held 
that historically, the Sloka metre is later than the Tristubh it need not 
necessarily impiy that these slokas are later than the Tristubh verses in the 
vatthu-gatha. The divergence in metre is perhaps additional proof of 
the difference in the periods of composition of the two parts. 

External Evidence: 

The story of the Buddha's nativity in the vatthu-gatha agrees in general 
with the versions in Lalitavistara, Mahdvastu, the Tibetan Dulva and the 
NidcSnakatha (Jatakd), but differs considerably in details. E. J. Thomas 
(Life of Buddha, pp. 38 ff.) has made a comparative study of this and no 
attempt is made here to go into any details. The verse-recension which 
follows the prose at Lai. 101 ff. has no connection whatsoever with the 
Ndlaka-discourse in Sn. but is merely a different version of the prose 
legend with enlargements and details which differ to some extent. Unlike 
the prose these verses bear no close resemblance to the vatthu-gatha. 
The points of interest in this account are: — 1. Asita's nephew is Nara- 
datta and not Nalaka as in Sn. 2. There are more miracles, but the 
Bodhisattva does not plant his feet on Asita's forehead as at J. I, 55. 3. 
Asita sees with divine eye — dibbacakkhu — the birth of the Buddha and 
informs his nephew of it declaring the only two courses of action open tosuch 

Pali Buddhist Review 3, 1 (1978) 


a being. 4. He takes his r.ephew with him to Kapilavastu and interviews 
Suddhodana and not the Sakyas as in Sn. 5. The thirty-two major 
characteristics are dealt with in great detail. 6. After his pronounce- 
ment that the new born babe would become the Buddha he returns to the 
asrama and advises Naradatta to follow the Buddha when the time comes. 
The version in Mvastu. which agrees with the vatthu-gatha is found quite 
apart from the iVa/o/ca-discourse, viz. Mvastu. II. 30 ff. and III, 382 
ff. (in both prose and verse respectively). Here Asita, the rsi from 
Ujjayini goes "to Kapilavastu with his pupils including Nalaka who later 
in the account is also called Narada. In details, the account is similar 
to that in Lai. Although the naimittikas, ' 'soothsayers" declare that the 
young prince would become a cakravartin, Asita is certain of his becoming 
the Buddha. There are a few other miracles such as the birth of 500 
each of girls, boys, male and female slaves, etc. simultaneous with that 
of the Bodhisattva. The verses that follow (pp. 33-43) deal more elabor- 
ately with the same incident. Asita advises Narada to practise the 
brahmacarya under the Buddha. The version at Mvastu. Ill, 382 ff. will 
be discussed later. It is also noteworthy that the Buddha's interlocutor 
here is called Nalaka Katyayana (p. 386) who on the advice of his father, 
the purohita to king Tonehara seeks ordination (by the formula, ehi 

According to the Tibetan account at Dulva, III. f. 461 ff. (Rockhill, Life 
of the Buddha, pp. 17 ff.) it is the statue of the yaksa Sakyavardhana that 
bows down at the child's feet and not the hermit as in J. The rsi Aklesa 
(=Asita— Rockhill) the dweller on the Sarvadhara mountain, with his 
nephew Nalada goes to see the infant Bodhisattva. He predicts the 
child's future and advises his nephew to enter the Sakyan order when the 
time comes. The Dulva further states that Nalada became known as 
Katyyayana among the 500 brahmins whom he joined at Benares, and 
that after his conversion by the Buddha he was called ' 'the great member 
of Katya's family". Thus, an attempt is made here to identify him with 
Mahakatyayana (cp. Mvastu. Ill, 386 Nalaka Katyayana). 

The Nidanakatha (J. I. 54 ff.) which decidedly shows signs of being 
much later than the BSk. accounts contains a great many details and 
abounds in miracles. The story, agrees fundamentally with the other 
versions. The name of the aged visitor is Kala-devala, 'Devala the 
Black' (Asita=Kala). He is a tapasa, 'a hermit' from Avanti Dakkhina- 
patha (Ujjeni, cp. Mvastu.) and not an rsi. As in the Dulva the exact 
time of the Buddha's enlightenment is stated (viz. after thirty-five years). 

Among other references to Asita in the Pali Canon is the mention of 
Asita-devala at M. II, 155 whom Malalasekera (s.v., DPPN) attempts to 


Pali Buddhist Review 3, I (1978) 

Pali Buddhist Review 3, 1 (1978) 


identify with Kaladevala an ascetic of Aranjara whose younger brother 
was Narada. In the Samodhana of the Indriva Jdtaka (J. Ill, 469) Kala- 
devala is identified with Kaccana which perhaps indicates a distant echo 
of Nalaka Katyayana in Mvastu. or Mahakatyayana of the Dulva (vide 
Rockhill, ibid, pp. 18, 45). There is also mention made of Narada of 
Aranjagiri in Majjhimappadesa, the younger brother of Kaladevala at J. 
Ill, 463 ff. and V, 133 ff. 

The relatively early existence of the account is evident from the diverse 
accounts which present a uniformity in essentials. The more elaborate 
versions in BSk. with all their embellishments are decidedly later than the 
vatthu-g&tha. Needless to say the Nidanakatha and the Dulva are much 
younger than the BSk. However, the general consistency suggests a 
common origin to all these accounts. Though it is said that there is no 
evidence to show that the legend itself was pre-Christian (Thomas, ibid.) 
this alone is no proof of its being so late as that. It may have had an 
independent existence long before it came to be fixed in some definite 
form in the vatthu-gatha. There is no reason to exclude the vatthu-gatha 
from the Sutta Nipata that was known to the author of Milp. (vide Milp. 
411. 414, etc.). Thus it is quite probable that this legend existed in pre- 
Christian times. At the same time it cannot be disputed that it was 
later than the sutta proper. 

Internal evidence has shown that it belongs to a younger stratum than 
the iVa/fl/ca-discourse. The fact that it has little bearing on the latter 
is further proof of its being an accretion at editorial hands, as was noted 
in the case of the vatthu-gatha of the Parayana Vagga. The independent 
accounts in Lai, Mvastu., Nidanakatha and Dulva, though they may be 
much later than Sn., further testify to the fact that the two parts of the 
sutta known as Nalaka Sutta in Sn. are in reality two independent poems 
differing in age, brought together at a subsequent date which, most 
probably, coincided with that of the final collation of Sn. 

The identity of Nalaka is made rather obscure by his being referred to 
as Nalaka Katyayana at Mvastu. Ill, 386 ff. Nalada Katyayana 's 
conversion at Dulva XI, f. 118 ff. (Rockhill, op. cit. 45-46) found quite 
independently of the story of the nativity shares something in common 
with the introduction to the Mauneya Sutra of Mvastu., for, the episode 
of the Naga Elapatra occurs in both of them. Yet, the individual in question 
is no other than Nalaka of the Pali sutta. Narada of J. Ill, 463 and V. 
133, ff. is quite distinct from Nalaka. Similarly it is doubtful whether 
Asitadevala of M. II, 155 who had a younger brother Narada was Asita 
of the vatthu-gatha. Perhaps the identification of Kaladevala at J. Ill, 
469 with Kaccana and the mention of Asitadevala may have been respon- 

sible for the name Kaladevala (of Ujjeni) in the Nidanakatha instead of 
Asita as in other versions" However, it is almost self-evident that with 
the passage of time and the spread of the story various confus.ons have 
set in as a result of the influence of foreign legends. 

Various attempts have been made by scholars to establish a connection 
between the nativity-legends in Buddhism and Christ unity. In the 
circumstances of the birth of Christ, Bunsen, 12 Seydel and Lilhe see 
an echo of the story of the Buddha's birth. C. F. Aiken 15 an Amencan 
theologian, sees in all these works "spurious evidence used to impugn 
the originality of the Gospels". E. J. Thomas (op. cit.) notes that Seydel, 
Edmunds and Pischel see in the story in Sn. the original story of Simeon 
(Luke, ii, 22-32) and that according to the latter 16 (Pischel) the differences 
between the two stories are less than their correspondence. Wmdisch, 
in Festschrift Kuhn has traced Asitadevala back to Brahmamcal literature. 
He regards it "not absolutely proved that the Simeon of St. Luke owes 
his existence to the Asita of the Buddhist legend". This subject is dis- 
cussed at length by Edmunds in his Buddhist and |»^ ««*■£»* 
Windisch in Buddha's Geburt, and by J. Kennedy mJRAS 1917 , pp. 209 ff., 
469 ff. Whatever similarities there are in these two stories Thomas has 
pointed out clearly the differences between them (ibid.). 

As noted earlier it is quite probable that this story was incorporated in 
the traditional accounts of the life of the Buddha in pre-Christian times. 
Thus any suggestion of a borrowing on the part of Chnst.amty or of a 
common origin prior to the birth of Chnstianity might cast serious 
aspersions on the originality and uniqueness of the legendary sections 
of the Gospel which many writers have endeavoured to maintain If 
there has been any borrowing at all both the Indo-Aryans in Madhyadesa 
and the Jews in ancient Israel may have probably drawn from a common 
source. However, the greater probability is that both stones may have 
originated independently of each other, and that they are merely parallel 
developments in the course of growth in the two rehgions. 

The Nalaka-discourse: . 

The sutta itself dealing with moneyya, as observed earlier, preceded the 
composition of the vatthu-gatha. The contents of the discourse with the 
emphasis on the conduct of a muni which points to a society of forest- 
dwelling ascetics, are ir^icativ^ofits^arh^o^ 

\t ThYma^Golama Buddha and the Gospel of Jesus Chnst, p. x.v. 
16. Pischel Leberi und Lebre des Buddha, p. 23 n. 

18 Pali Buddhist Review 3, I (1978) 

and modes of conduct for the monk, and therefore belongs to that category 
of suttas m Sn. de S1 g nated as the ' 'muni-class ' '. There is a higher ethical 

hTH derlyin f o the S " tta; and tWS iS mUCh more P^nounced than even 
in the Mum and Sammdparibbdjaniya Suttas. It has the same tone as the 
Khaggavisana Sutta and agrees with its ethical values which have a soecial 
reference to the bhikkhu. p 

No detailed observations need be made on the language, style and 
metre of the sutta. A few casual remarks, however, have to be made on 
the ideology On its own merits, the sutta recommends itself as an early 
poem, for all mternal evidence clearly indicates it. The language is old 
and preserves several archaic forms many of which are poetic. Unlike 
the vatthu gatha it contains no late forms. The ideas in the sutta share 
r<nY^ mm ° n r th contem P° rar y ^dian thought. Neumann (Reden. 
p. 504 ff.) has made a comprehensive study of this giving many parallels. 
The stanzas Sn 702, 703,705, 711, 712, 713, 714, 720-722 are outstanding 
examples of ideas common to all literature of the period, though the 
thought throughout the sutta is more characteristic of an ascetic sect 
However, a distinctive Buddhistic strain runs through the whole poem' 
Yog,c P ract,ce S are mentioned in Sn. 716, and in the above mentioned 
stanzas are to be seen echoes of the Brahmanas and the Upanisads. 

Although there are several accounts of the Buddha's nativity in Pali 
^ , i , hterature > the onl * vision which bears a close resemblance to the 
Nalaka-discomst in Sn. is to be found at Mvastu. Ill, 386 ff. Practically 
the whole discourse occurs in similar words with a few changes which 
however, do not show much divergence from the Pali. The order of the 
24 stanzas in Mvastu. is different from that of the 25 stanzas in Sn There 
is nothing in the BSk. version corresponding to Sn. 718,- and the stanzas 
paralle to Sn. 709 and 714 bear only a vague resemblance to them. Sn 
/06 is slightly expanded in BSk. whileSn. 708 ab , 707 ab and 708 cd 709 cd respec 
lively form two stanzas and Sn. 719 a has no parallel at all' The other 
noteworthy changes are :~Sn. 702 d / /ksanto c&numato bhava, Sn 71 V 
a-gammallasadya, Sn. 71 l d payutam //prepsutdm Sn. 708 b abhiharavcll 
abhiraksayc ■ Sn. 715* visatdf/sarita; Sn. IWI/evam mauneyam upesyasi 
ana bn. 714 // na par am dvigunayati na pi caiva gundy ati. 

The close resemblance these two versions bear to each other suggests 
that they are but two recensions of the same discourse. It may be pro- 
bable that the Pali version is older than the BSk., but it cannot be supposed 
that the latter is based on the former. The only justifiable conclusion 
is that they had a common origin. Besides this there are several Money- 
ya Suttas in the Pali Canon. Anguttara Nikaya I, 273 contains a short 
sutta dealing with the three moneyyas, kaya, vaci and mano. entitled 

Pali Buddhist Review 3, J (1978) 


Moneyya Sutta. The abstention from the three akusalas arising from 
these three sources is termed moneyya and the sutta is concluded with the 

kayamunim vacdmunim celomunim nasavam 
munim moneyyasam pannam ahu sabbappahdyinam. 

Itivuttaka III, hi, 8 (It. 56) contains a more concise version of the same 
sutta. It mentions the three moneyyas and concludes with the same 
stanza as at A. I, 273, but with the last pdda altered into ahu ninhatapa- 
pakam. The ten abstentions are not enumerated here. Sangiti Sutta 
(D. Ill, 220) merely mentions the three moneyyas together with other 
groups of threes. 

Although the Ndlaka Sutta neither specifies the three moneyyas under 
kaya, vaci and mano, nor enumerates them as the ten abstentions all that 
and much more is implied in it. The discourse covers a wider range than 
the limited scope of the sutta at A. I, 273. In addition to the abstentions 
(Sn. 704-706) there are positive injunctions on the mode of conduct of a 
monk. The absence of a well defined classification, and the emphasis 
which still lay on the life of the muni, the forest-living recluse, are also 
indicative of the iVa/a&fl-discourse being anterior to the suttas mentioned 

It has already been observed that Asoka's Moneya-sute was the Ndlaka 
SuttaFBR 1, 3p. 138. Mrs. Rhys Davids (Manual, pp. 312-314) identifies the 
fifth dhammapaliydya with it. III,iii, 8 (It. 367 is evidently a misprint for 
//. No. 67), and Winternitz (op.cit.l, 607) accepts it. The alternative name 
of this sutta was the Moneyya Sutta (Mvastu. Mauneya), and it is most 
improbable that Asoka would have meant either It. 56 or A. I, 273 by his 
Moneya-sute, for there is nothing remarkable about these two pieces 
whereas Ndlaka Sutta has every claim to it. The thought and sentiments 
in it are so lofty that it had to be included in the list with its companion 
poem the Muni Sutta. Besides this, the musical Sloka metre may have 
also been responsible for its popularity, for as stated earlier (ibid.) a sutta 
in verse would naturally have been preferred to a passage in prose. Hence 
Mookerji's suggestion (Asoka, p. 118n) that by Moneya-sute was meant 
the Ndlaka Sutta should be accepted." 


17. The suggestion that Asoka's Moneya-sute meant the Thera-and Therl-Gathci 
(Max Walleser) is untenable. 


Pali Buddhist Review 3, 2 (1978) 

That brahmin, hearing these words, asked both for forgiveness and for 
the Going-forth and practising the development of loving-kindness was 
taught in this way by Brahmadatta Thera: 

If anger should arise in you 

reflect on the Simile of the Saw, b 

if craving for flavours should arise 

remember the Son's flesh Simile. 7 

If your mind runs craving 

pleasures and existences 

bind it quickly with mindfulness 

as a beast found eating corn. 8 (441-446) 

6. Kakacilpama Sutta, Majjhima Nikaya 21 . 

7. Vide Related Collection (Samyutta Nikaya) II, 63 (and Wheel Nos. 105-6, The 
Four Nutriments of Life). 

8. Simile of the Lute— vide Related Collection IV, 205. 


N. A. Jayawickrama 

Some Suttas from the Atthaka Vagga 

Kama Sutta 

The Kama Sutta which appears at the head of the Atthaka Vagga presents 
many problems. The four suttas consisting of eight stanzas each and 
called Atthakas by the compilers follow the Kama Sutta. Judging from 
the evidence furnished by Pali sources alone, the natural inference is that 
these are the true Atthakas and the vagga including the rest of the suttas 
was named after them. The possibility of an alternative explanation 
has been suggested earlier. 

As said above these four Atthakas form one group and the rest of the 
suttas form the other group (or groups). It is not possible to say whether 
these four suttas formed the foundation on which the superstructure of 
the rest of the vagga was bi.ilt, or v hether they formed an ornamental 
carving on the already exist. :g edifice of the vagga, finally providing those 
characteristics which suppli id the name to the vagga which it now bears. 
Linguistic evidence may perhaps famish a clue to its solution. 

The stanzas are examined individually below: — 

Sn. 766; the cognate use in kamam kamay amanassa (v. i. kamay anassd) 
is old and poetic and is of restricted usage in subsequent literature ; ce 
as a conditional conjunctive as in 767 a is restricted to gaiha. The 
ellipsis in pada 6 is metri causa. The pada has the ring of an old gatha, 
specially the emphatic particle used. 

Sn. 767; The medial ppr. kamayanassa is old gatha from Vedic origin. 
The gender o?kama is uncertain in this sutta; kama alternates with kamani 
(77 1 b ) . Of the 5 instances the word occurs in the sutta it is decidedly 
rnasc. at 763 a and 769 a and probably masc. in this stanza (though 
traditional grammarians recognise an — a form in the neat pi.). It may 
be either masc. or neut. at Sn. 766 a (ace. sg.), but is neuter at 771 & 
though the pronoun referring to it seems to recognize it as masc. (but te 
is occasionally used as neut. pi. nom. and ace). The verb, rupvati dates 
back to an "r-dialect" in Vedic. (CI. 3k. has l-\/'up, Jump). It is most 
frequently u;ed in this phrase (c, . S. I. 198; TM, 967; Sn. 331, etc.) and 
met with it- later literature except in grammatical works in v'hich 

is not 

ami etymology is suggested for ritpa. 


Pali Buddhist Review 3, 2 (1978) 

Sn. 768; pada. is a shorter Vedic inst. sg. and sin is the Vedic ace. sg. 
(historical) as opposed to Pali sirasatn or siram. The only term, with a 
doubtful exception of abala (77(f), to which, a technical significance 
could be attached is visattikam. 1 It occurs 8 times in Sn. viz. Sn. 333 c , 
768 c , 857 d , 1053 d =1054 d =1067 d =1085 d =1087 d . Where the evolu- 
tion of the idea is concerned this line appears the oldest of them all, for 
its use here is non-technical. 

Sn. 769; The collective dvandva cpds. are old. The word porisam 
belongs to an old stratum (v. I. parisam, cp. Sk. paurusam). It appears 
as a collective neuter. The contracted form thiyo is historical; and the 
word puthu is met with in the old language (cp. Jataka verse). 

Sn. 770; According to the Corny, the word abala is technical (tanhg), 
but its use in any technical sense elsewhere is not met with. The phrase 
abala va nam ballyanti could best be translated as, ' 'being weak themselves 
(i.e. kama) they overpower him" (cp. Chalmers' translation). A similar 
phrase is seen at J. IV, 84 (verse), vata ballyanti (cp. Pv. II, 6i). Line c, 
cp. Dh. 1, 77zl, 735. The simile of the ship is continued in 771 d . The 
v. I. silva may be compared with Dh. 369 {^fsrll to depend on). All these 
instances show that the language of the Kama Sutta is necessarily very old. 

The metre of the poem is different from that of many other parts of the 
vagga (including the four Atthakas). It is in the Sloka metre like Nos. 
7, 10, 15 and part of 16, i.e. {Sn. 814-823, 848-861, 935-954, 955-962). 
The majority of the suttas is in tristubh viz. Nos. 2-5, 8, 9, 11, 12,13, and 
part of 16 of the vagga (i.e. Sn. 772-779, 780-787, 788-795, 796-803, 824-833 
835-847, 862-877, 878-894, 895-914, 963-975. The Jam Sutta (Sn. 804-813) 
and Sn. 834 are in Vaitallya. It has been emphasised earlier that metre 
is no safe guide to the date of a poem in Pali, for, not all tristubhs in Pali 
date back to a pre-Sloka period. The lack of uniformity in metre in 
the whole vagga and the fact that essentially most of its suttas are linguisti- 
cally old lead to the hypothesis that it was formed from already existing 
older material. TheKama Sutta is one such instance, and the only conclusion 
that could be drawn (from the analogy of the Ratana Sutta) is that it was 
one of the last suttas to enter the vagga. On this account it cannot be 
proved late, for it may have had an independent existence prior to its 
introduction here which itself had taken place at a very early date. 

The theme of this sutta is a very popular one in Pali. Instances where 
monks and laymen are advised to give up kama are too numerous and 
therefore need not be mentioned here. Although there are many passages 
in the Canon dwelling on this topic, there is hardly any section which 

1. cp. Ardhamagadhi (Jaina) visottiya=Sk visrotasika. 

Pali Buddhist Review 3, 2 (1978) 


bears a resemblance to this sutta. J. IV, 167-172 contains a set of 9 
gathas of which the first is identical with the first gatha here, but the other 
8 are different — though in tone and theme they are similar. Again Dh A . 
Ill, 284 commenting on Dh. 216, 2 contains a passage similar to this sutta 
but the words and the tone are quite different. (Also vide PTS s.v. kama). 
The other instances where gathas of this sutta are found repeated are 
either quotations or examples for commenting viz. Sn. 766-768 are com- 
mented at Nett. 69, Sn. 766 ab quoted at PsA. 50, Vism. 378; Sn. 767-771 
commented at Nett. 6; Sn. 767 quoted at SA. I, 32, Vism. 576; and Sn. 
769 quoted at UdA. 120. Sn. 768 is common with 77;/, 457, 769 ab with 
J.IV, 240, 771 d with Dh. 123, J. IV, 173, and 771 b with Dh. 369. The 
lines and padas that are common to Sn. and other works cannot be 
established as borrowings. 


The Atthakas 

The four suttas following the Kama Sutta consist of eight stanzas each 
and hence are called Atthakas in Pali. It has already been shown that 
this fact has led Pali compilers to designate these suttas by this name and 
take a further step to exte- ; the name to the whole vagga. The theme 
of the first of these ' 'octaves" is closely related to that of the Kama Sutta. 
In fact the Guhatthaka Sutta appears as a continuation of it and deals 
with the same question more comprehensively on a psycho ethical basis. 
The psychological concepts such zsguha and mohana (Sn. 772) are common 
to other schools of contemporary Indian thought. The term satto has 
a special significance, i.e. attached to the guha. The psychological basis 
of this sutta is further seen in terms like mamayita, amama (Sn. Ill) 
ubhosu antesu (778), ditthasutesu (778), and sahnam (119). As parallelisms 
with the previous sutta Sn. 779 a maybe compared with Sn. 771, c parigga- 
hesu (779 b ) with Sn. 769, and Sn. 779'' with Sn. 770 while appamatta may 
be said to refer to the yogic ideal. 

On account of the similarity of the themes of the Kama and Guhatthaka 

Suttas it may be argued that the Kama Sutta was placed in front of the 

Guhatthaka aiming at an arrangement in accordance with subject matter. 

This, however, has not met with much success, for suttas 6 and 7 of this 

vagga bear an appreciable resemblance to suttas 1 and 2 in this respect. 

If these two were placed immediately after the Gidiatthaka the four 

Atthakas would not have remained as a group. In the same way the 

2. Tanhaya jdyati soko, tanhaya jdyati bhayam, 

tanhaya vippamutiassa natthi soko kuto bhayam. 


Pali Buddhist Review 5, 2 {1978) 

subject matter of suttas 3, 4 and 5 resembles that of 8, 12, 13 and the 
discourse in 9. Instead of these suttas following one after the other they 
occur in three separate groups showing on the one hand the incomplete- 
ness of the classification, and the partial adherence to a method of 
arrangement according to external form, on the other. 

The three suttas following Guhatthaka deal with the various aspects of 
one and the same theme. They indicate the Buddha's attitude to philoso- 
phical speculation. The Dutthatthaka points out the position of a muni 
who is beyond all censure and has become steadfast by casting off (\/dhu) 
all philosophical views (ditthi). The Suddhatthaka ridicules the notion of 
attaining purity (suddhi) through metaphysical speculation and emphasises 
the importance of remaining aloof from biases and limitations. The 
Paramatthaka declares that philosophical disputation should be given up 
and that a true and steadfast sage needs no philosophical views to lean on. 

In all these suttas, as well as in Nos. 8, 9, 12 and 13 and numerous other 
old suttas of the Canon the futility of metaphysical speculation is em- 
phasised. The Buddha's attitude towards the subject is made evident 
in them. From a historical examination of the dominant ideas in them 
it could be inferred that they represent a very early stratum in Buddhist 
thought. The excessive indulgence in metaphysical subtleties of later 
Buddhism, specially that of Ma.ha.ya.nic schools affords a clear contrast to 
the ideas and sentiments of these suttas. The main theme is the relin- 
quishment of philosophical dogmas but other references to fundamental 
tenets of early Buddhism (e.g. Sn. 790 c , 792 cd , 793% 794 cd , 800 c , 801 ab , 
803 ab , etc.) are clearly indicative of the spirit of early Buddhism that these 
suttas breathe. Disputation is condemned. It is not a knowledge of 
metaphysics that is sought after, but a life of selfless wandering free from 
attachments to the states of being (777 cfl ) and unmeasured by sense- 
impressions (778 d ). The essentials on which early emphasis lay are sum- 
marised in Sn. 779. The muni is not sullied by "graspings" (pariggaha), 
he crosses the "flood" by the realisation of sanha (SnA. 518 ndmarupa), 
has uprooted the dart, wanders diligently and yearns for neither world. 
Again, the muni has no khila (stubbornness, Sn. 780 c ), he is serene and 
released and does not proclaim his attainments (783 ab ). He has no 
theories which he has evolved and fabricated (784 a ) and is not one whose 
peace is dependent on mutability (784 d ). He is a dhona (he who has cast 
off everything ) and is independent in every way (Sn. 786). He has 
reached that state when he has no views either to approve of or disprove 
any dogmas (Sn. 787 e ). 3 

3. SnA. 523 comments on attain nirattain as: attadifthi va ucchedaditfhi va natthi. 

Pali Buddhist Review 3, 2 (1978) 49 

Many of the terms used in this section to describe the muni (both 
epithets and phrases) have a philosophical tone. They are in some 
instances technical, but the majority of them were yet to develop into 
technical expressions with definite values. The Commentary attempts 
to explain khila as rdgadi khila which shows a definite development of the 
termbythe time of the compilation of SnA. (cp. Sn. 973 b , 212°, 477°540 d , 
1059 d and 1 147 d ). The only instances where it has a technical significance 
are Sn. 540 and 1147. A gradual process of crystallisation is to be seen 
in ussada at Sn. 783 d . Its philosophical import is evident in all the 
occurrences of the term in Sn. viz. 515 d , 855 d , 624 b and 920 d . The 
usual seven ussada are given at Nd\, 72. The term dhona in the Atthaka 
Vagga is used with reference to the shaking off of ditthi (Sn. 786°, 813 a 
and 834 c ); but at Sn. 351 b it is an epithet of the Buddha. Nd\, 77 
explains it as pannd while the Commentator gives the interpretation of his 
day. The wider application seen usually in Commentaries (viz. SnA. 542, 
J. Ill, 160) is not to be met with in Sn. for it is solely used to signify the 
abandonment of ditthi. Equally abstract and semi-technical in use is the 
term upaya (also anupaya, Sn. 787 ab , 797 c , 786 d ) but it has not found 
much in subsequent literature. One of its cpds. rupupaya occurs at S. Ill, 
53, etc. cp. Sn.A. 522, tanhaditthiupayanamdvinnamabhavenaanupayo. . sam 
dvinnam bhavena upaya cp. SnA. 558 upagantabbatthen upayam rupddisu 
ekam pi dhammam upeyya. In both these instances the Commentator 
interprets from the level of his day, and the context does not justify the 
inclusion of tanha in SnA. and Nd\. 82. 

The two phrases kuppa-paticca-santim (Sn. 784 d ) and at tarn nirattam 
are also interesting. The Commentator has seen too deep into the 
meaning of kuppa-paticca-santi when after a long comment he explains 
it as; tan ca anisamsam tan ca kuppataya ca patic asamuppannataya ca 
sammutisantatdyd ca k.p.s.sankhdtam ditthim nissito va hoti (cp. Nd 1 , 74-75). 
The phrase "characterising the peace which is dependent on mutability" 
describes the santi of him who sees virtues in himself on account of his 
speculative theories. E. M. Hare translates it as ' 'Calm on quaking built' '. 
It is in fact no technical term. The Corny, is again seen giving the inter- 
pretation of its day to attam nirattam where it speaks of attaditthi and 
ncchedaditthi (SnA. 523 and Ndl, 82) taking atta to mean atman and 
nirattam the BSk. nairdtmyam which is a later development, (attam — 
attam). The universe of discourse here is ditthi (philosophical views); 
and hence attam and nirattam cannot refer to anything else but the accept- 
ance or rejection of ditthi. In the light of the subsequent elaboration of 
the anatta doctrine which was a sine qua non in the earlier teaching, this 
word has undergone a complete transformation. 


Pali Buddhist Review 3, 2 (1978) 

That the term upadhi (sopadhlka Sn. 789°) has a definite connotation 
even in Sn. can be seen from the various instances in which it occurs. Sn. 
728 makes it quite clear. 4 Also cp. Sn. 1050 cd 1051, 33 c , 34 c , 546 a 
572 a , 364 a , 33 d , 34 d , 642 b 371°, 1057 b , 1083 b and 992 f . All these 
occurrences show that the term has undergone a definite crystallisation, 
and there is no doubt that the concept belonged to the earliest stratum 
of Buddhist thought. 

The phrase, ditthe sute silavate mule va (in what is seen and heard, in 
ascetic practices and holy vows and in what is cognised— Sn. 790") is a 
curious combination of functions of the senses on the one hand and 
external practices on the other. In this context dittha, suta and muta 
(muta from y/mon, l.E *\/mn) imply the sights, sounds and other undefined 
sense impressions respectively which are considered auspicious and pure 
(cp. Sn. 790 a antlato suddhim aha: and Nd\, 8711. : SnA. 527 comments, 
mute ca uppannena micchananend). Both Nd. and SnA. are not clear 
about muta. This idea occurs no less than 20 times in Sn. in similar 
words viz. Sn. 790 ab (=797 b =887 a ), 793 ab =914 ab (793 b =A. II, 25), 
798 cd (c=S. I. 203), 797 ab ( 790 b , 887 a ), 887 ab , 910 ab , 1079 bc (= 
1080 bo , 1081 bc ), 1082 cd (=1083 cd ), 788 b (=789 a ), 802 ab , 897 d , 778 d 
(=250 d ). In all these instances the psychological basis of the reference 
to sense impressions is hidden by the nature of the context which either 
introduces or implies silabbata along with it. It is clear that all these 
references do not merely speak of the functions of the sense organs which 
produce the result but mention the result itself. Yet, Sn. 1086 a and 
1122 cd seem to bring out the psychological aspect clearly viz. idha dittha- 
suta-muta-vinndtesu (in things that are seen, heard, sensed and perceived), 

na tuyham adittham asutam-mutam va 
atho avihndtam kiiicanam atthi loke 

(there is nothing that is not seen, heard or sensed or else not perceived — ■ 
cognised — by you in this world) as at D. Ill, 134, 232, It. 121. In Sn. 
897, dittha and suta the functions of the two primary senses only are 
mentioned as at Sn, 778 d , 250 d , 1079 b 1080 b and 1081 b , although they 

4. Upadhlniddna pabhavanti dukkha 
ye ke ci lokasmhri anekariipd; 
yo ve avidvd upadhim karoti 
punappunam dukkham upeti rnando; 
tasmd pajdnam upadhim na kayird 
dukkhassa jdtippabhavdnupassi. 

(Those diverse forms of sorrow which prevail in the world arise basing their origin 
on the material substratum. Indeed, the indolent fool who nurtures his material 
substratum repeatedly brings himself to sorrow. Therefore should he who discerns 
and comprehends the origin of the arising of sorrow not accumulate his substratum). 

Pali Buddhist Review 3, 2 (1978) 


are intrinsically connected with silabbata. The same idea is expressed 
at Sn. 839 a and 1078 a as well as Sn. 840 ab //839 ab . Although these 
references are similar to each other in meaning, dittha and suta (and muta) 
in combination with silavatani mentioned or implied, are essentially 
different from dittha-suta-muta-vinndta in their fundamentals. The latter 
has a more universal application and is primarily meant to describe the 
functions of the senses (muta representing those of the three senses not 
mentioned under sight and sound, and vinnata that of manas). 

The early Buddhist emphasis on the detachment from both punna and 
papa is seen at Sn. 790°. It is aptly described as attanjaho in line d i.e. 
abandoning whatever is "grasped" (dtta cp. 800 a , 787° not as at Nd\, 
90 attaditthijaho nor SnA. 527 attaditthiya yassakassaci va gahanassa 
pahinatta attanjaho, both of which being interpretations of a later level). 
The idea of ' 'crossing over" which is so frequent in the early Pali literature 
(ogham \/tr or param \/ gam; vide the introduction to the Pdrayana, 
PBR, 1, 3, p. 146, is found here as at Dh. 412, 370, Thl, 633, Sn. 212, 473, 
etc. in its special reference to sangam (attachment). A sahhasatto (Sn. 
792) is one who is led by his senses; lit. "attached to percepts". Both 
SnA. 527 and Nd\. 93 speak of him as the opposite of vidvd. The idea 
of a simatiga brdhmana is common to all stages of Buddhism cp. also 
tadi, etc. 

A comparison and analysis of all these ideas shows that they belong to 
the earliest strata of Buddhism. As pointed out earlier, some of them 
are in an early stage of development while others have undergone a certain 
degree of crystallisation. It is also noteworthy that some of these con- 
cepts as upayo which have not undergone any development here are 
scarcely found in later works or other works which may claim equal 
antiquity with Sn. On the other hand, elaborate theories and extensive 
treatises are to be found in later literature with regard to the more import- 
ant of these concepts which developed fully under favourable conditions. 
A mere study of the ideology of these Atthakas and a careful examination 
of where the emphasis is laid in the poems reveal their very antiquity. 

Linguistic data which form a very important factor for the determination 
of the age of the ballads confirms what has been arrived at by means of 
other criteria. In fact, in the case of these poems, linguistic data con- 
clusively establish their antiquity. It is very significant that all the old 
forms in these suttas point to some Vedic dialect of Pali rather than to the 
standard canonical Pali. The language in general reflects a form of early 
Pali. It is not proposed to examine every stanza individually. However, a 
brief survey will make the position clear. In this short section of 32 
stanzas there are four full Vedic double forms with a dialectical (perhaps 


Pali Buddhist Review 3, 2 (1978) 

Magadhl) influence viz. cutdse Sn. 774 d , avitatanhdse 77 6 d , sitase 79 1% 
and paticchitdse 803 b . There are 9 ppr. forms ending in -am and-ana viz. 
tittham 772 b , yfi^am 773 d , caram 779°, abhijanam 788 c , vadanam 789 d , 
pamuncam 79 l d , paribbasano 796 a , anupadiyano 800 a , and anadiydnam 
802°. There are some words which are restricted to the Atthaka Vagga 
only e.g. paribbasano at SW. 796, 878, 880, and 895, three of which occur 
at the beginning of a ^wWa (i.e. except 880). There are archaic verbal 
forms as jahha Sn. 775 b , pdva 782 bd , and pa va 789 d . The middle base 
kubba-is preferred to kar-\ kubba occurs at Sn. III d , 778?, 781 c , 790 d , 
and 794 d ; faur-at 796 b and fcsr-at 800 b and in pvrekkharonti at 803 a , 794 a 
and purakkhata 784 b . There are a few other unusual verbal or secondary 
forms as suppahdyd 772 d , duppamuncd, anhamokkhd 773 b , avaddniya 
774 b , parinnd, accaycyya 781% svdtivattd 785 a , niccheyya 785 b , 801 d , 
vikappayeyya 793 d , 802 d , nmayeyya 798 d , kappayeyya 799 a and a«w- 
paneyya 799°. There are also two medial optative 3 sg. forms, sikkhctha 
775 a , and mahhetha 799 d which are characteristically ga/Zaa forms. 

The syntax too points to an old idiom. There are at least 10 instances 
of the construction with the historical locative of relation in varying shades 
of meaning viz. at Sn. 772 a , 774 a , 776 cd , 777 d , 779 b , 783 b , 785 b0 , 786 b , 
787 a , and 793 a . All these are sufficient data to prove the antiquity of 
this section of the Atthaka Vagga. 


Jara Sutta 

This sutta consists of 10 stanzas in vaitdllya metre. The only other 
vaitdllya verse in the vagga is Sn. 834. The theme of the poem is the tran siency 
and impermanence of life. One is advised to leave the household life 
"seeing that no worldly possessions are eternal and that everything is in a 
state of flux". Emulating the sages — munayo Sn. 809 — the wise man is 
exhorted not to form any egoistic attachment to anything conceived as 
"one's own" since everything is left behind at death — Sn. 806. Death 
leaves behind only the memory of the dead. 

The above ideas in Sn. 804-809 closely conform to the title of the sutta. 
Although the last four stanzas — Sn. 810-813 — appear somewhat foreign 
to the sutta under its present title, all of them except the last stanza are 
connected with Sn. 809; and they fit in with the genera.! theme on account 
of the similarity of ideas. Sn. 810 d is the logical extension of 805 d , and 
similarly 81 1° is closely associated with 809 a . The sage is called a 
dh.ona in the concluding stanza; and in this respect 813 a may be compared 

Pali Buddhist Review 3, 2 (1978) 


with 786 ab , line c with 824 b and the whole stanza with 790, 793-795, 914, 
etc. The line d, na hi so rajjati no virajjati, breathes the same air as the 
concluding lines of the Suddhatthaka, 

na rdgardgl na virdgardtto 
tassidha n'atthi par am uggahltam. 

The uniformity of metre suggests that the poem as a whole dates back 
to the same period. The theme of the poem which is in praise of the 
mwm'-ideal is common with other poems of great antiquity in Sn. Suffici- 
ent has been said already on this topic and it not proposed to discuss it 

The language of the sutta calls for particular attention. In discussing 
the stanzas individually any striking points in ideology and doctrine will 
be pointed out. Sn. 804 the ablative in oram vassasata represents the 
old idiom, miyyati (lines bd) = impersonal medial cp. Sk. mriyate. An 
absolutive in adverbial function is seen in aticca (l.c); cp. upaddya gacchati, 
samddaya rakkhati, etc. The form jarasd can be explained in two ways; 1. 
inst. sg. of a noun jaras, an extension of the -as declension (besides jara 
f. and jara m. or n.); 2. -sd adverbial suffix from the analogy of the adverbial 
inst. of -as nouns. The whole stanza is rather elliptical. Grassmann, 
Worterbuch zum Rgveda, points out 6 examples from the Vedas where 
jara (s) is masc. cp. also the inst. at Rv. X. 85. Thus this is an old form 
in Sn. going back to a Vedic dialect. Sn. 805 cp. Mbh. XII, 805 and Asia 
Prajha Fdramitd 254. I. a, cp. 777 a , 809 b . Of the 17 occurrences of 
mamdyati or its verbal derivatives, as many as 9 are found in the Atthaka 
Vagga. Mamayita is clearly the earlier word signifying egoism. The 
word atta is not so frequently used in this sense in Sn. The opposite idea 
amama occurs 5 times in Sn. whereas anatta occurs only twice viz. anattani 
(756 a ) and anattagarahl (913 d ). Of these two instances only anattani 
(756 a ) has some connection with amama, but as this occurs in the relatively 
late Dvayatanupassana Sutta it may be surmised that amama stood for 
anatta and mamdtta for the parallel idea atta or attaditthi. The evidence 
at hand is insufficient to establish whether this really was the germ of the 
more comprehensive anatta theory of Buddhism. Another word which 
is in popular use in this vagga ispariggaha (five out of the seven occurrences 
in Sn. are in the Atthaka Vagga— viz. 393 c , 470 b , 809°, 805 b , 871 b , 872 b , 
and 779 b ). The use of this word is necessitated by the subject matter, 
and it is semi-technical. It is evident that the central theme is the transi- 
ency of life and the impermanence of worldly possessions. The title 
Amama Sutta or Anagdriya Sutta would equally fit the poem, for specially 
the last few stanzas emphasise this aspect. The cpd. in l.c appears to be 


Pali Buddhist Review 3, 2 (1978) 

an expression of popular origin. Sn. 807 a supinena (with samprasarana 
and svarabhakti). The consonant group .srar-more frequently undergoes 
samprasarana than assimilation in Sn . There are six instances o f samprasa- 
rana (viz. supati 110°, supitena 33 l b , supina 293 d , 360 b , 807 a , and 927 a ) 
as contrasted with one instance of assimilation (sond 675°) and one 
instance of svarabhakti and consonantal hardening (supdnd 201 a ). Metrical 
exigencies may have promoted this tendency, but the scarcity of assimilated 
forms may be significant as pointing to a particular dialect. I.e. piyayitam 
cp. mamayitam. Sn. 808 d : akkheyyam has the appearance of a deliberate 
pun (i.e. from d+^khyd or a-\-\/ksi). SnA, 543 comments, Namamattam 
eva tu avasissati. (The mere name remains); Nd 1, 127, Rupagatam, 
vedandgatam, sahndgatam, sankhdragatam, vihndhagatam pahiyati. . 
namam evdvasissati. Akkheyyanti akkhdtum, kathitum, bhanitum, dlpayi- 
tum, voharitunti, namam evdvasissati akkheyyam. (All that pertains to 
the fivefold aggregates perishes. . only the name remains. Akkheyyam 
means to name, to speak, to address, to elucidate and to employ in usage; 
and only the name remains to be spoken of (or understood). Also vide 
PTS s.v. d+\/khya. It is quite probable that this is a gerund from a+V 
ksi (vide PaninI, VI, 1, 81), i.e. ksayya>kheyya, cp. sayyd>seyya. The 
Brhadarayaka has the same idea (Brh. Ill, 2, 12), Ydjnavalkyeti hovdea, 
yatrdyam puruso mriyate, kim enam na jahdtt ti, ndmety anantam ndma 
(Yajnavalkya said: when a man dies what is it that he does not give up? — 
It is the name for it is everlasting). Also cp. Maitri Upanisad II, 4, 6, 28 
ananto' ksayyah (endless and imperishable) which seem to suggest that 
Pali akkheyya may be from a+\/ksi. 

The Atthaka Vagga contains 9 out of the 1 1 references to the word 
jantu in Sn. The parallel word which is more frequently used in Pali 
is satta 5 (10 times in Sn.) and jantu has almost gone out of use in later Pali 
(s.v., PTS). It is only in one doubtful instance that satta (as referring 
to creature, being) occurs in the Atthaka Vagga i.e. satto guhdyam bahund- 
bhicchanno—Sn. Ill 3 - (from^srjl). 

Sn. 810. In line a is found one of the numerous instances where the 
word bhikkhu is used in the same connotation as muni. The word bhavana 
in line d is apparently a synonym for bhava. It occurs again at Sn. 685 b , 
937° and once in prose. Sn. 811, 812: The points of interest in these two 
stanzas are the similes in 81 l d and 812 ab , which are in fact the same simile 
stated in different words. Along with 812b may be cited, 

5. cp. Rafthapala Sutta (M.II. IV, 2): satto pana gacchati yena kammam where 
sat ta= jantu. 

Pali Buddhist Review 3, 2 (1978) 55 

padumam va toyena alippamanam (Sn. 71° cp. 213°), 
nopalippati toyena, toy ena padumam yatha (Th\, 70r d ), 
pundarlkam yatha vagga toye na upalippati (Sn. 517 ab ), 
pundarlkam va toyena sahkhdre nopalippati (Th\, 1180 ab ): 

and with Sn. 71 l a and 812 a the following: 
vari pokkharapatte va (Dh. 401, It. 84, Sn. 625), 
udabindu va pokkhare (Th\, 665 b ), 
bhikkhu yatha pokkhare vdribindu (Sn. 392 d ), 
udabindu va pokkhard (Dh. 336 d Th\, 401 d ) and 
vdribindu va (M. Ill, 300, J.VI, 595). 

Sn. 811 ab is significant as showing the detachment of a muni (cp. 813 d 
already discussed). 

The general tone of the Jard Sutta is archaic. It definitely represents 
the old ga^fl-language. The thought in the sutta like that in many other 
old pieces is representative of the times. The Salla Sutta may be cited as 
a close parallel to this poem in ideology. The only difference is that it 
deals primarily with death while amama is emphasised in the Jard Sutta. 
All the available internal evidence is in support of its early composition, 
and it is quite probable that it is as old as the four Atthakas. 

Magandiya Sutta 

The Magandiya Sutta is a dialogue of 13 stanzas in tristubh metre. The 
context of the sutta is the occasion of Magandiya's futile effort to give his 
daughter away in marriage to the Buddha. The story is narrated in 
detail at SnA. 542-544. There is also a dialogue between the Brahmin 
Makandika and the Buddha at Divy. 519-520 which is incorporated in a 
prose and verse mixed narrative (Divy. 515-521). The two narratives 
at Divy. and SnA. agree in general, but differ in details. In the Pali 
Commentary the Buddha foresees the good fortune of the brahmin 
Magandiya and his wife to attain arahatship (arahatta phatupanissaya) 
and contrives to meet the brahmin. In Divy. it is a chance meeting. The 
names ofMakandika'swife and daughter are given as Sakati and Anupama 
respectively in Divy., but the Pali gives only the feminine of the family 
name as Magandiya. Divy. contains a full description of the conversation 
between the brahmin and his wife about their future son-in-law, and 
introduces a new character, and old man who eagerly offers to marry 
Anupama when the Buddha refuses her hand. Makandika refuses his 
offer and he vomits hot blood and dies. It states nothing further of the 


Pali Buddhist Review 3, 2 (1978) 

Makandikas while the Pali mentions the attainment of arahatship of both 
husband and wife (SnA. 518). 

The dialogue at Divy. 519-520 which consists of 5 stanzas shows some 
resemblance to a few corresponding stanzas in Sn. viz. st. 1 roughly 
corresponds to Sn. 835, st. 2 has some bearing on Sn. 836, and stanzas 
4 and 5 together are somewhat parallel to Sn. 845. The ideas in st. 4 ab 
are similar to those at Sn. 845 ab , though they are not identical. The 
simile at stt. 4 cd and 5 ab is the same as at Sn. 845 cd . The idea expressed 
at Sn. 845 cf is found at st. 5 od . The only difference between them is that 
in Divy. these two stanzas are uttered by the Buddha about himself, 
whereas in Sn. it is the muni who is described. 

Again, Sn. 835 speaks of the three daughters of Mara as actual persons 
and not as mere personifications of ideas in an allegorical representation as 
at Sn. 436 (Padhana Suttd) where arati and tanha are mentioned as the 
second and fourth senas of Mara. Raga is to be identified with kama 
in Sn. 436. In the Magandiya Statu the three daughters of Mara are actual 
persons. In the Divyavadana their names are not mentioned, and the 
stanza runs, 

drsta maya Marasuta hi vipra, trsna Na me na'pi tat ha ratisca 
chando na me kamagunesu kaschit, tasmad imam mutrapurlsapurnam. 

Although the daughters are alluded to, tanha (trsna) and rati are qualities 
mentioned along with kamagunesu chandas (cp. methunasmim chando). 
Judging from this it is very difficult to state definitely which version 
preserves the older tradition. In both cases the personification seems to 
have been long forgotten and Mara is conceived as an actual being who 
had three daughters. 

Judging from the abruptness of the change of topic and the transition 
from one subject to another in Sn. 836 cd it may be argued that Sn. 835, 
836 are versifications of an old prose introduction. It is also a plausible 
explanation that the basis of the sutta is the Buddha's encounter with 
Magandiya. This is common to both versions, and without falling into 
the error of presuming that the BSk. version is older than the Pali, on 
account of its brevity, a common source may be assigned to both. From 
the evidence of Divy. any suggestion that the two opening stanzas were 
foreign to the sutta is untenable. Moreover, there is no difference in 
metre and language between Sn. 835, and 836 and the rest of the poem, 
and by no means are these two stanzas an interpolation of a compiler. 

Yet, a closer comparison of the two versions shows that the main theme 
of the Pali poem is not found in BSk. The Magandiya Sutta praises 

Pali Buddhist Review 3, 2 (1978) 


the muni who does not enter into disputes and has inward peace whereas 
the central topic in BSk. is Makandika's quest for a son-in-law. The 
first three stanzas deal directly with it and the last two are given as the 
Buddha's own words of self-praise. These stanzas lack the detached and 
impersonal refined note struck in the stanzas of the Pali in which the 
Buddha praises the muni when invited to speak about himself (Sn. 836 cd ). 
It is evident that Sn. 837-847 can stand as an independent sutta without 
the two opening stanzas (Sn. 835-836). This leads to the inference that 
probably the Pali sutta represents a fusion of two independent ballads of 
which Divy. 515-520 forms only one component, affording a parallel to 
Sn. 835-836. 




The Vatthugdtha and Commentarial literature state that Ajita was a 
disciple of the brahmin Bavarl, although according to AA. I, 184 he was 
Bavarl's nephew. Theragatha (Th\, 20) mentions an Elder Ajita who had, 
in a former birth, offered a kapittha fruit to the Buddha Vipassi. The 
Commentary on this stanza (77zl A. I, 78) refers to him as the son of the 
assessor (agghapaniya) of the King of Kosala. Apadana No. 509 (Ap. II, 
449) also speaks of a Kapitthaphaladayaka Thera (cp. 77z 1, 20), but there 
is another Apadana of Ajita the pupil of Bavarl (No. 397-Ap. I, 335). There 
is no attempt made in the Commentaries to identify Ajita, the Kapittha- 
phaladayaka with Ajita of the Parayana. The subsequent growth of the 
Bavari-episode in connection with the Parayana (PBR, 1, 3, p. 146), the 
antiquity of the Parayana itself (ibid), and the vagueness with, which some 
Commentators refer to it, 6 are additional testimony to the fact that the 
Ajita of the Puccha and the Ajita of Th\, 20 (cp. Ap. II, 449) are two 
different persons. The Vatthu-gathas refer to the sixteen questioners 
as, sissa solasa brahmana. The name Ajita need not necessarily be that 
of a brahmin («-y7-ta=unconquered); and it suits a ksatriya equally well. 
It is significant that these sixteen are spoken of as ayasma Ajito, ayasma 
Punnako etc. in the pucch&s. They address the Buddha on equal terms 
as marisa as do all ksatriyas and the warrior gcds of the Hindus (Sakka, 

6. AA. IV, 35: Addhateyyagathasataparimartarrt Parayanasuttam (P. Sutta which 
consists of 250 stanzas); but the entire vagga with its Prologue and Epilogue contains 
only 274 stanzas, pucchas alone being 92 stanzas. The Parayana is called a sutta here. 
Nd2 also refers to some pucchas as suttas and pafihas. 


Pali Buddhist Review 3, 2 (1978) 

etc.). It is only in the titles of the pucchas that they are called manava 
(the text of the prologue and the epilogue is of no consequence for obvious 
reasons). The word manava, which often designates a young brahmin 
is no conclusive proof of these men being brahmins. Some of the names 
are decidedly ksatriya; e.g. Ajita Bhadravudha, the names Nanda and 
Hemaka are doubtful, and Pirigiya and Mogharaja are most probably 
nicknames of ksatriya origin. Neumann (Reden 546) sees in the name 
Bavarl a representative of the famous Katyayana school of the White 
Yajurveda (Badarl). He says that among the manavas there are seven 
other Yajurveda priests of whom four belong to the White Yajurveda. He 
also mentions a still older Badarl of the Black Yajurveda to whom reference 
is made in the Baudhayana-grhyasutra (I, 7). Even if his suggestion is 
accepted there are still nine others who have to be proven brahmins. 
Moreover, a name like Dhotaka, which Neumann had in mind (his seven 
Yajurveda priests are not enumerated) is a fitting name for a disciple of 
the Buddha (■s/dhu, dhunati, to shake off, to purge, etc. cp. the concept 
dhona which is often used as an epithet of the muni in Sn.). Likewise 
Mettagu, Upaslva, Ajita and Tissametteyya 7 are very suitable names for 
the Buddha's disciples. 

The first question asked by Ajita is very far-reaching. 8 On the one 
hand it could be interpreted empirically to mean only the external objects 
of the world, on the other it implies Ajita's premonition of world-sorrow. 
The Buddha in his reply alludes to the First Truth: dukkham assa mahabb- 
hayam. In his next question Ajita goes a step beyond the answer and 
anticipates further. This clearly shows that Buddha's interlocutor was 
a person with a considerable previous metaphysical training. The second 
question is asked in a fashion that makes it possible to illustrate indirectly 
the Four Noble Truths. Because Ajita himself has some idea of the 
misery inherent in the world he is eager to know by what means it could 
be checked. Following the Buddha's reply (Sn. 1035) he shows his 
desire to know how sati, pahha and the individual namarupa cease to exist. 
Here the question hints at nirodha (or perhaps upekkha as well), and in 
the reply the very word nirodha is used. That Ajita thinks clearly ahead 
and anticipates the replies is evident from his question in Sn. 1036. 

These questions are far too brilliant to be those of an insignificant 
disciple of a brahmin from the less-known and least-brahmanised zone 
of the Dakkhi-napatha which even during the time of the compilation of 

7. There is another Tissametteyya in Atthakavagga 7. 

8. Ajitapuccha is commented at Netti, pp. 70-72, under Sodhanahdra Netti, III. 13. 
It states that Buddha's replies were in the form of sodhand and not arambha (on his own 
initiative) viz...//' panhe.Ji Bhagavd padam sodheti no ca arambham. Ajitapuccha 
is again commented at great length at Netti, 10-21. 

Pali Buddhist Review 3, 2 (1978) 


the Baudhaya-nagrhyasiitra was considered unfit for brahmins (Baudh. 
V, 15). Further, the trend of thought in these questions compares rather 
closely with the monistic principles of the Upanisads. The macrocosmic 
Brahman, identified with Atman, the world-soul, gives place to the 
microcosmic Atman which again is identified with the macrocosm. Though 
no such philosophical subtleties are in evidence here the progress from 
world-sorrow to namarupa is reminiscent of the Upanisads. The picture 
of Ajita in the puccha is not that of a typical brahmin youth but that of a 
mendicant initiated into the Upanisadic way of thinking. One would 
not be far wrong to conjecture that since the ksatriya seers were the 
custodians of Upanisadic lore and as Ajita's mode of thinking resembles 
their's that he was a ksatriya belonging to an Upanisadic school. The 
very fact that his name sounds like that of a ksatriya or of a sage, 'The 
Unconquered', is no conclusive proof of his ksatriya origin. 

Linguistic and other Internal Evidence 

The sutta generally bears the appearance of an old piece. There is a 
preponderance of the use of the particle su (or ssu) as an emphatic interro- 
gative. This is a general characteristic of many old dialogue-ballads in 
which the interlocutor continually asks questions. Among forms which 
may be assigned to a dialectical stratum are; jappa (Sn. 1033c) which is 
not confined to gatha and marisa (1038d) found equally well in prose. 
A double Vedic form with the MagadhI ending is to be seen at Sn. 1038a, 
sahkhatadhammdse. The sandhi ki'ssa (1032c) is probably dialectical 
viz. kirn assa>ki assa (nasalised i)>ki' 'ssayki' 'ssa cp. Pv. Ill, 5, 6. kVssa 
vatam kirn pana brahmacariyam where ki perhaps contains an original 
nasalised vowel. In both these instances kVssa cannot be identified 
with the interrogative pronoun kissa in the oft-repeated formula tarn kissa 
hetu. Also cp. Pv. II, 6, 1, Utthehi Kanhekl sesi; the corresponding passage 
to it at J. IV, 79 reads as kim sesi. 

The other peculiarities are more of a purely grammatical nature, yet 
pointing to an old stratum of Pali; e.g. short abl. singulars veviccha, and 
pamada (Sn. 1033b), pithiyyare (1034d, 1035d) of Vedic origin (cp. Geiger 
§ 122) with consonantal hardening. 

The term sola (1034, 1035) is used in the sense of defilements such as 
tanha (SnA. 586). Of similar application is sota at Sn. 355 

Acchecchi tanham idha namarupe (ti Bhagavd) 
Kanhassa sotam digharattanusayitam 

(He has completely cut off the desire for name-and- form— individual exist- 
ence here, the stream of Kanha which had remained for long). Existence 


Pali Buddhist Review 3, 2 (1978) 

is often spoken of as a stream; e.g. bhavasota at Sn. 736b, S.I, 15, IV, 128, 
etc. It is considered a positive attainment to rid one's self of this sota; 
e.g. chinnasota Sn. 715b, and also sotam chindati M.I, 226. The flux of 
mind is also a stream, vihndnasota D. Ill, 105, etc.; and the Noble Eight- 
fold Path is called a stream (sota) at S.V. 347. The terminology of Ajita 
is allied to Buddhist terminology though at first sight the term appears to 
be used in a connotation different from that in Buddhism. 

Style calls for no attention. The puccha is in sloka metre (anustubh), 
and metrical irregularities are few viz. an even quarter at 1037', a short 
pdda at 1036 a , and extra-syllabic padas at 1033 ab . 

The Other Pucchas 

Like Ajita, the other 15 manavas too have questions to ask the Buddha. 
Tissametteyya wishes to know of the mahapurisa who is unperturbed and 
perfectly contented. Punnaka asks the Buddha about the efficacy of 
sacrifice and the reasons why men offer sacrifices. The Buddha replies 
that it is all futile and that it would not enable one to overcome birth and 
decay. Then he expresses his desire to know of them who have transcend- 
ed birth and decay. Mettagu asks the Buddha the reason for the existence 
of suffering in this world and the method by which the wise cross the 
stream of birth, decay, sorrow and lamentation. Dhotaka invites the 
Buddha to preach to him to enable him to train himself for his release 
and remove ail his doubts. Upasiva requests the Buddha to give him an 
arammana (means, object) by which he may cross the Flood (ogha). 
Nanda asks whether it is knowledge or the mode of living that characterises 
a muni. Hs also wishes to find out whether those who profess metaphysical 
theories have overcome birth and decay. Hemaka tells the Buddha that 
he took no delight in the theories of the Vitandavadins, and requests him 
to preach to him the dhamma by which he may transcend 'this sinful 
bent'. Todeyya asks the Buddha about the nature of the emancipation 
of him who has no craving, is free from lust, and has overcome doubt. 
Kappa asks him of the island-refuge from the formidable stream con- 
fronting the mortal subject to decay and death. Jatukanni requests the 
Buddha to tell him of the santipada and to preach to him the dhamma to 
help him to leave behind birth and decay. Bhadravudha praises the 
Buddha and requests him to preach the dhamma to all. Udaya wishes 
the Buddha to declare to him the deliverance by transcendental knowledge 
and the destruction of ignorance. Posala asks the Buddha about the 
state of knowledge of the person whose consciousness of form is extinct, 

Pali Buddhist Review 3, 2 (1978) 


who has cast off corporeal form and perceives internal and external 
'nothingness'. Mogharaja asks the Buddha how one should look upon 
the world so that Mara may not 'see' one. Piiigiya like Jatukanni asks 
the Buddha to preach the dhamma in order to leave behind birth and 

A striking feature of many of the pucchas is the eagerness of the 
questioners to listen to the Buddha. Some of them come with special 
problems that had confronted them. Their earnestness is seen in Sn. 1061, 
1097, 1120. Nanda's question gives the Buddha the opportunity of 
stressing the superiority of a moral life (cp. 1070c). He declares that 
speculative knowledge leads one nowhere. This is in contrast to con- 
temporary Brahmanism where Upanisadic seers begin to emphasise the 
importance of knowledge (jnana) for the attainment of Brahman. Vijja 
(knowledge) in Buddhism in some aspects is allied to jnana, yet the Buddha 
is seen consistently (o reject metaphysical speculation (cp. Atthaka Vagga, 

In reply to Upasiva 's request the Buddha gives a short survey of the 
essence of vimokkha. This puccha appears the most abstruse in the whole 
vagga. The concentrated ideas in it are highly philosophical and bear 
the tone of the more systematised passages of the Anguttara of similar 
import. It perhaps represents in germinal form the doctrines further 
dealt with in the Anguttara imd Samyutta Nikdyas and carried to a degree 
of perfection in the later Abhidhamma Pitaka. 

The arammana which Buddha gives Upasiva is based on akincanna (cp. 
na kincid anyat). He has to cross the ogha by obtaining the release brought 
about by sanndvimokkha (cp. sanndvedayitanirodha). Then only dees 
a muni 'go beyond reckoning' and obtain his release from nama (ndma- 
kaya), for rupa is eliminated at the stage of akincanna. Here is a brief 
reference to the kaya theory of the Nikdyas. The Potthapada Sutta in the 
Digha mentions the various kdyas as conceived by the divergent schools of 
animistic beliefs of the existence of a soul. The term rupakdya occurs 
at S. Ill, 59 and nama-kdya is that which corresponds to the entities 
designated as nama in the division of the fivefold aggregates. Here is also 
to be seen a distant echo of the kosa theory of the Upanisads. There is 
nothing quite close to this in the Nikdyas, but the significant metaphor 
asim kosiyd pabbdheyya (as one would draw the sword from the scabbard 
D. I, 77) seems to suggest an early connection of the same ideas. 

The central ideas of the pucchas are discussed in the general remarks 
on the Pdrdyana Vagga (PBR, 1, 3, p. 146). All the concepts in the vagga 
are doubtless very old. The passages of philosophical import do not 
show much growth. The occurrence of the terms vinndna (1055), 1073, 



Pali Buddhist Review 3, 2 (1978) 

1110, 1111 and 1037, namarupa (1036, 1100), namakaya (1074) and nama 
and rupa (1073), akincanna (1070, 1071, 1072, 1115) has already been 
touched upon. The terms kincana (1098, 1099, 1 104) and akincana (1059, 
1091, 1094) are of no direct philosophical import. Vihnata (1086) in the 
phrase dittha-suta-muta-nhnatesu is a term common to passages dealing 
with sense-perception. The notion of going beyond samkham (1074), 
pamanam (1076), kappam (1 101) agrees with the central theme of 'going 
beyond'. Like the suttas of the Atthaka Vagga the pucchas denounce 
disputation (takka 1084, kathamkatha 1088, 1089) and philosophical 
praise ofthe Buddha (1043, 1049, 1063, 1069, 1073, 1090, 1101, 1105, 1112, 
etc.). He is called samantacakkha at 1063c, 1069c, 1073b, 1090d. The 
other frequent epithets are aneja (1 1 12, 1 101, 1043), vedagu (1049, 1059), 
bhavitatta (1049) and oghatiga (1096). None of these epithets appears 
extravagant and all could be ascribed to an early period. The dhamma 
is spoken of as amtiha (not based on hearsay— 1053). Santi is to be 
experienced in this world itself (1066). The terms itihltiham and itthab- 
hdva also occur. No attempt is made here to discuss other data from the 
language of these pucchas, for both language and metre show signs of 
antiquity and agree in the main with the suttas of the Atthaka Vagga. 

It is to be observed that only one (Punnaka) out of these sixteen men 
asks a question about sacrifice, a thing which played a very prominent 
part in the lives of all the brahmins of the age. This question is the only 
justification to infer that Punnaka was a brahmin; though in itself it is no 
conclusive proof. It has already been emphasised that some of the 
questions asked, definitely show that most of them have had a philosophical 
training in some system or other. It is quite probable that they may 
have belonged to some sects of sramanas or ajivakas which cannot be 
easily identified on account of the scanty evidence at hand. 

The only mention in the Apadana, a considerably late work, of the 
celebrated Bavari of the Vatthugatha, with special reference to these 
manavas, is made at Ap. II, 487 (Mogharaja), Ap. II, 342 (Mettagu) and 
Ap. II, 357 (Todeyya). It was stated earlier that the fact that some of the 
names are brahmin-names does not necessarily prove that the questioners 
were brahmins. Dhotaka in praising the Buddha calls him a brahmana 
and in the same stanza addresses him as Sakya (1063). At 1065 he calls 
him brahme (voc). This presents no difficulty when the new significance 
attached to the word brahmana is taken into account (cp. Dh. Brahmana 
Vagga). The main purpose of these questions is to find out a solution to 
birth and decay and not the settlement of the differences between the 
theories of these interlocutors and Buddha's teaching, for none of them 

Pali Buddhist Review 3, 2 (1978) 


comes to the Buddha as a disputant. All this evidence points to the late- 
ness of the Bavari episode as compared with the pucchas and shows the 
absence of any justification for the late tradition that these manavas were 

The Apadana and the Sixteen Manavas 

The only other source in which these manavas are mentioned in a 
manner worth noting is the Apadana. Only eleven out of the sixteen are 
specifically mentioned, viz. Ajita: Ap. No. 397 (I, 335), Tissametteyya: 
No. 398 (II, 339), Punnaka: No. 399 (II, 341), Mettagu: No. 400 (II, 342), 
Dhotaka: No. 401 (II, 343), Upaslva: No. 402 (II, 345), Nanda: No. 403 
(Ap. II, 350), Hemaka: No. 404 (II, 351), Todeyya: No. 405 (II, 354), 
Jatukanni (ka): No. 406 (II, 357), and Mogharaja: Nos. 35, 537 (I, 87; II, 
486). There is no trace whatsoever, in the Apadana, of Kappa who 
should have been mentioned after No. 405, of Posala or of Pingiya. There 
is the story of one Udena occurring in the Apadana immediately after 
Jatukannika (i.e. No. 407. Ap. II, 362). Following this comes the 
Apadana of Bhaddall (No. 408. Ap. II, 365). Although the names appear 
somewhat similar the stories yield no clue for the identification 
of Udena with Udaya and Bhaddall with Bhadravudha. Moreover, the 
order in which these two stories occur is the inverse of that of the two 
corresponding pucchas. Even in the case of the eight Apadanas in which 
there is no mention of Bavari (i.e. except Todeyya, Mettagu and Mogha- 
raja) the text affords no positive evidence of a connection. 

Further, Udena's Apadana is the last number of the 41st (Metteyya) 
Vagga and Bhaddall 's opens the next chapter which is known by that 
name. This may be overlooked if there was any positive evidence of a 
connection, for Ajita's Apadana ends the 40th (Pilinda) Vagga and there- 
fore precedes the Tissametteyya Apadana. The division of the Apadana 
into vaggas being arbitrary and artificial, it is evident that Ap. Nos. 397- 
405 are meant to correspond to the nine manavas from Ajita to Todeyya. 
The tenth, Kappa is omitted, and the eleventh, Jatukanni occurs as No. 
406. Then comes Mogharaja the fifteenth manava for whom there are 
two Apadanas. Sn. 1117 is quoted at Ap. No. 537,25; and Sn. 1118-1119 
at Ap. No. 537, 26-28. Though there are differences in details the two 
stories are practically the same. The fact that the Mogharaja Apadana 
is so far away from the last story which has a bearing on the manavas 
(Jatukanni) hardly sheds any light on Sn. on account of the lateness of Ap. 

References in other works 

From the nature of the questions and answers in the pucchas it is to be 
inferred that the manavas entered the Order. This is stated in the late 


Pali Buddhist Review 3, 2 (1978) 

Epilogue (Sn. 1128). Yet it is rather disappointing to see that Th\ is 
silent about most of them. It has already been shown that Ajita at Thl , 
20 is not the same as Ajita of Sn. Similarly, Punna (Th\, 70), Nanda 
(Thl, 157-158), Posiya (Thl, 34) nor any one of the three Tissas in Thl, 
(39, 97, 153-154) shows any connection with the men bearing similar 
names in Sn. It is also highly improbable that Bhaddali (Thl, 275-277) 
or Bhadda (77il, 473-479) and Udayl (Thl, 689-704) have any connection 
with Bhadravudha and Udaya. The degree of improbability is less in the 
case of Kappa of Thl, 567-576, though no direct evidence is forthcoming. 

On the other hand, it is quite probable that Mogharaja of Thl, 207-208 
is the same as Mogharaja in Sn. In fact he is the most frequently mention- 
ed person out of all these sixteen manavas. It has already been stated that 
he is mentioned twice in the Apadana (I, 87, II, 486) and once in the Thera- 
gatha (Thl, 207-208). Samyutta, 1, 23 contains two stanzas, one by Mogha- 
raja and the other the Buddha's reply, which are not found either in Sn. 
or Thl. It may have been quite probable that the original Mogharaja- 
puccha was longer than what is now handed down in Sn. It is also pro- 
bable that the Samyutta quotes from another recension of the Mogharaja- 
puccha which is now lost. The quotation found at Milp, 412 of a saying 
by Mogharaja cannot be traced either in Sn. or Thl. It is probable that 
the source from which it was taken was known to the author of Milp. 
and was subsequently lost. The nature of these passages does not permit 
the inference that they belonged to another Mogharaja. This corrobo- 
rates what has already been noticed in the case of the two Apadana stories. 

It is not proposed to give an analysis of the linguistic data. The few 
remarks made earlier show to some extent the antiquity of these poems. 
All the evidence from external sources points to the fact that Mogharaja 
was a prominent member of the community. The evidence from the 
Samyutta and Milp. does not help to establish the anteriority of the poem 
in Sn. to those respective works. It is quite probable that Sn. preserves 
only a fragment of a longer dialogue; and that the Samyutta and Sn. 
are complementary to each other in this respect. 

(to be concluded) 


According to the Pali Commentaries : Text and Introductory Essay 

Bhikkhu Bodhi* 


In the whole of Buddhist literature the epithets of the Buddha are probably 
as numerous as the sands of the Ganges, but among these many names 
and titles none cuts deeper or takes in more meaning than the word 
"Tathagata". It is the epithet the Buddha uses with greatest frequency 
in reference to himself, a fact of no small importance. It is never employ- 
ed by others as a direct address for the Buddha, and only rarely as a third- 
person designation, as though its use was a privilege reserved for the 
Master alone. The context in which the Buddha uses it further heightens 
its significance. When he speaks about himself as a particular individual 
with a private background of experience, the Buddha generally employs 
the first-person pronoun "I" (aharii). But when, in contrast, he speaks 
about himself in his other, supra-individual role — as the discoverer and 
teacher of the path to deliverance — he drops this personal mode of speech 
and refers to himself instead indirectly as ' 'the Tathagata. ' ' The very 
appearance of the word in the suttas thus signals a shift in perspective: a 
momentary flash from the tight analytical principles composing the 
discourse to the vast spacey backdrop against which its exposition unfolds. 
We see this already at the outset of the Buddha's career. When 
the newly enlightened Master first approaches the five ascetics in 
the Deer Park at Benares, he begins his teaching mission by announcing: 
"The Tathagata, monks, is a Holy One, a perfectly enlightened Buddha." 
And so it is, day in and day out, all the way through the remaining 45 
years of his ministry, right down to the Pari-nibbdna. Whenever his self- 
reference draws him out from the web of particulars in which his life is 
set to reveal him in the full breadth and majesty of his stature, time and 
again the words come: ' 'The Tathagata, bhikkhus, " 

In recognition of its pre-eminence among the Master's epithets, the 
early Buddhist teachers and their successors have applied their wisdom 
and erudition to fathoming the multiple implications of this suggestive 
word. Their tradition of exegesis, transmitted and elaborated from one 
generation to the next, has reached its standard form in the detailed 
explanation set down by the great dcariya, Buddhaghosa Thera, in his 
polished editions of the ancient commentaries. In what follows we 


N. A. Jayawickrama 

§1. In the series of contributions to the PBR concluding with the present 
article, an attempt has been made at an analysis of the Pali Sutta Nipata 
in the light of Higher Criticism. Various factors of the latter were 
grouped under the categories of 'criteria' which enabled us to estimate 
the single suttas as well as groups of them from the angle of literary, 
doctrinal and linguistic development. Special attention was paid, in 
the application of these criteria, to view the Sutta Nipata under the aspect 
of historical development, illustrating its gradual growth. Wherever 
possible, external evidence was adduced in the historical interpretation 
of the data furnished by internal sources. 

A study of methodology was one of the main concerns of this under- 
taking. Copious examples of each category have been given to illustrate 
and (as far as possible) prove the propositions; and special attention 
has been paid to samples of textual interpretation. Exegesis was both 
synthetic and analytic and the foundation on which it was built is the 
historical background of Indian (Hindu and Brahmanic) ideas around 
and prior to the time of Asoka. 

By 'growth' is implied the gradual formation of a separate anthology 
called the Sutta Nipata by the incorporation of suttas belonging to diverse 

For purposes of investigation this work has been divided into four 
parts. Part I (Introductory). The chief criteria (which fall under the 
heads of linguistic, metrical and literary evidence, doctrinal developments, 
growth of ideas and external and indirect evidence) employed in the 
examination of the ballads and other poems of the Sutta Nipata were 
discussed in PBR 1, 2. The remarks (ibid) dealing with the title 'Sutta 
Nipata' and its form and contents are also of an introductory nature, 
Part II. A brief study of the five Vaggas of the Sutta Nipata was made 
in PBR 1, 3. Special attention has been paid to explain the present 
arrangement of the suttas in their respective vaggas. A few topics of 
general importance such as the seven dhammapaliyayas of Asoka's 
Bhabra Edict, the Chinese version of the Arthapadam (Atthaka Vagga), 
the title 'Atthaka Vagga' and the relation of the vatthugatha to the 
pucchds of the Parayana were also discussed there. Part III. The 
contributions in PBR 2, 1 to 3, 2 dealt with the analysis of a few suttas 

Pali Buddhist Review 3, 3 (1978) 


representative of the various types of poems of the Sutta Nipata, with the 
aid of the criteria detailed in PBR 1, 2. The poems examined were: 
three ballads from the Uraga Vagga, (Uraga, Khaggavisana and Muni 
Suttas), five suttas of popular character (viz. Parabhava, Vasala, Mahgala, 
Metta and Ratana), the Yakkha-ballads (Hemavata, Alavaka and Suciloma 
Suttas), the Pastoral-ballads, Dhaniya and Kasibharadvaja Suttas, the 
narrative-ballads, Pabbajja Padhana and Nalaka Suttas, suttas from the 
A tthaka Vagga and the Pucchas of the Parayana. Now, Part IV is devo ted 
for general observations and conclusions. 

The composition of the majority of these poems can be assigned to the 
period 400-300 B.C. On the evidence available, it is clear that individual 
suttas have to be taken on their own merits though to some extent 
particular types of suttas can be vaguely generalised as belonging to 
distinct strata. The results which this investigation points 1o fall under 
the following headings:— (1) an early nucleus of a more or less 
floating tradition; (2) several intermediate redactions incorporating 
suttas drawn from the Buddha-legend and Buddha-worship; (3) a final 
redaction made for the purpose of propagating the Buddhist faith through 
its ecclesiastic representative, the Sarigha. 

§2. In the analysis of the suttas (he. cit.), with the aid of the criteria 
detailed in PBR 1, 2, a few general tendencies have been observed. Many 
of the poems, on linguistic grounds, appear to be old; but it is not always 
that the evidence from other sources is in support of this. Generally 
speaking the poems of the Atthaka and Parayana Vaggas and many of 
the pieces of the Uraga Vagga, in addition to those poems which can be 
termed as the 'Muni-ballads', represent the oldest stratum in the Sutta 
Nipata. Before finally enumerating the results which this investigation 
has led to, a short synthesis of the various data will perhaps be helpful 
in obtaining a better perspective of the Sutta Nipata as viewed from the 
angle of Higher Criticism. This synthesis will be mainly devoted to 
some aspects of linguistic data; and the characteristics of the later com- 
positions can be inferred thereby. No special attention will be paid to 
the nominal forms and the few remarks made in isolated instances on the 
vocabulary are deemed sufficient. Yet, some interesting nominal forms 
have been touched upon . On the other hand, a study of the verb and the 
verbal derivatives sheds further light on the Sutta Nipata as a whole. 
However, dialectical variations, Vedic characteristics, style and metre 
will again be touched upon in passing. Sufficient has been said in the 
individual suttas taken up for analysis on the doctrinal developments; 
and comparisons with similar poems (in Sn.) and classes of ballads have 
to some extent shown the general trends in Sn. A short survey will be 


Pali Buddhist Review 3, 3 (1978) 

made of the epithets and other terms used in Sn. to describe the muni and 
the Buddha. 

§3. The Sutta Nipata is rich in verbal forms and shows a very wide 
range. There are over 2,364 finite verbs in Sn. leaving aside variant 
readings and p.p.p's. used in a finite sense. Of special interest are the 
verbs in the Optative and Imperative Moods and the Aorist and Future 

Optative. — The most favoured inflexion for the opt. in Sn. is -e for 3 sg. 
In all, about 192 forms end in -e, but as many as 31 of them are either 
causals or medials (of the 10th class) with the element -ay- (i.e. -aye); e.g. 
haraye; ddiye, cintaye, etc. A small number of these optatives in -e 
belongs to the first and second person singular; e.g. sikkhe, Sn. 1061a, 
1062d, etc. Next in number come the forms in -eyya. Considering 
the fact that this is the most popular conjugational element for opt. in 
Pali, (vide Geiger, § 128) it is surprising to note that there are only 135 
such forms. Geiger (ibid.) does not class these forms as very early, for 
-eyya is a generalization of the Sk. -eya which underwent universal applic- 
ation in Pali. Of the 135 forms, no less than 115 are 3 sg. Another 
inflexion used frequently is a for 3rd sg. (Sk..-dt); 71 times. Its use however, 
is restricted to a few roots, e.g.\/jna, 33 times, \/as, 27 times (assa and 
siyd), \/kr (kayira) 7 times, etc. The opt. 3 sg. in -etha occurs 57 times, 
and the 2 sg. in-etho only once (Sn. 833c). Opt. 1 sg. in -am occurs 6 times 
(vijannam, Sn. 482a, 1020d, 1022e, 1065b, 1090c, 1097dj. The first pi. 
in -mu or -ma occurs 8 times (jdnemu, Sn. 76d, 559f, 999ad, janiydma 
873d, namassemu 995e, sikkhema, 89c and 32b v.l. carema-se). The 3rd 
pi. in -u or -um (I I Sk. -uh) occurs 10 times. It i? evident that these forms 
are old. As seen above, some of the forms ending in -um have also the 
element -eyy- which can be directly traced to Vedic (and Sk.) -ey. Besides 
those belonging to the type kathayeyya (Sn. 980d) which are accepted as 
old (Geiger § 139), the majority of the 135 forms in -eyy cannot be classed 
with the later types enumerated by Geiger (ibid), viz. 1 sg. manteyyam 
(Sn. p. 103), 2 sg. aroceyyasi (M. II. 210), dhareyyasi (Milp. 47). 3 sg. 
jaleyya (M. II. 203) and dasseyya (Milp. 47). 

Imperative. — The imp. in Sn. can be tabulated as follows: — 
2 sg. in -a, 43; in -hi, 110; in -ssu, 23; 

2 pi. in -tha, 40; in -vho, 3; 

3 sg. in -ta, 16; 
3 pi. in -ntu, 6. 

All these forms are historical in varying degrees, but are used in all 
stages of the language, and therefore are of no great value. 

Pali Buddhist Review 3, 3(1978) 


Aorist.— Parasmaipada: Following Geiger 's classification (Geiger, 
159) 37 verbs can be said to belong to type I (33, aor, 3 sg., 3, aor. 2 sg., 1,' 
aor. 3 pi), 63 to type II (40, aor. 3 sg., 2, aor. 2 sg., 5, aor. 1 sg., 13, aor! 
3 pi., 3, aor. 1 pi.), 90 to type III (72, aor. 3 sg., 5, aor. 2 sg., 3, aor. 1 sg., 
10, aor. 3 pi.) and 119 to type IV (67, aor. 3 sg., 5, aor. 2 sg., 7, aor. 1 sg., 
38, aor. 3 pi. 2, aor. 1 pi.,) which make a total of 309. Atmane- 
pada: There are 18 A'pada aor. forms. Of them 11 belong to type II 
(3 sg.), 4 to type 11(1 sg.), 2 to type IV(3sg.) and 1 to type IV (1 pi.)— vide 
Geiger, ibid. Among these forms are a few augmentless aoris!s. Some 
original pf. forms can still be distinguished, eg. aha, vedi, etc. The 
impf., lost in Pali, is represented by type II and the characteristics of the 
impf. are preserved in many of them. 

Future Tense. The sign of the future tense conjugation i-ssa- and the 
terminations -mi, -mo, -si, -tha, -ti, -nti are used in 46 future tense 
verbs. A form with issamase occurs once (Sn. 814d). The future 1. sg. 
-issam occurs 9 times, and ssam (without the connecting vowel, i) is used 
twice with thematic roots (upessam, Sn. 29c, and sossam 494c). The other 
historical forms—anupadassati (da-sya-ti, Sn. 983b). kahasi (kar-sya-si, 
-ss-> -h- 427d, 428d) ganchisi (*gam'sya- i/y, 665d), dakkhiti (draksyoti, 
i/y, 909a), dakkhinti (p. 14), pavakkhami (vaksyami, 701c, 963d, 1050b), 
bhasihi (bhas-ya- i/y, analogical 719a), dakkhinti (cp. dakkhinti 28c) and 
sagghasi (Vsak. 834d). This brings a total of 72 future forms. 

§4. The verbal derivatives too show an old phase of the language. 
The Agent Noun, Absolutive, Present Participle, and the Future Passive 
Participle will be discussed below. The Past Participle Passive will not 
be touched upon as it yields no definite irformation. The Infinitives 
and other forms of Vedic or dialectical origin will also be mentioned. 

Agent Noun.— There are 21 agent nouns in Sn. distributed in the follow- 
ing manner in the five vaggas:— 8, 1, 6, 4 and 2 respectively. 

Absolutive (Gerund).— There are 389 absolutives in the whole of Sn. 
gathas. Of them as many as 187 end in -ya, i.e. 1 1 1 formed with vowel- 
ending roots in Pali, in addition to 2with-aya>a(parihna, Sn. 779a, palik- 
hannd 968b), 66 with consonant-ending roots and assimilation, in addition 
to 3 with -yy- (pappuyya, Sn. 593b, 829d, 482d) and 7 forms with the 
svarabhakti vowel (a-r-i). Of these 187 forms, 185 contain prefixes 
conforming to the Sk. rules. The two forms without prefixes are:— 
gahaya, Sn. 79 Id, and yaciya 295b. Of the others, there are 117 formed 
with -tva, directly from the root. There are 8 forms with -tva assimilated 
(labh+tva>laddha). There are 26 forms formed directly from the base. 
Of the 48 forms with -tvana, 45 are formed directly from the root, two 


Pali Buddhist Review 3, 3 (1978) 

with the base and one form with assimilation (laddhana, Sn. 67c). The 
form datthu occurs 3 times (Sn. 424b, 68 Id, 1098b). Of these 202 as 
many as 36 forms contain prefixes. It is clear that the majority of these 
forms go back to a very early stratum in the language. 

Present Participle. — In all, there are 350 present participial forms. Of 
them as many as 139 are medial (107 contain the sufhx mdna and 32 -ana, 
both of which are highly archaic and go back to Vedic dialects). The 
occasional pronominal terminations of ppr. forms with -mana (e.g. Sn. 
434a, and manassa, 7 times) do not indicate that they were late, for in 
early Sk. too -smin and -sya are the terminations for the sg. of loc. and 
gen. respectively for ppr. medials in-mana. Of the rest of the 211 ppr. 
forms as many as 191 are historical. The total lack of forms like gacchan- 
tassa and gacchantam and the exclusive use of forms like gacchato and 
gacchatam for the gen. sg. and pi. respectively show that the ppr. too 
represents an old stage of the language. The 20 forms which cannot be 
considered equally old consist of 17 nom. singulars in -onto and 3 loc. 
singulars in -ante. But in Pali these endings came to be fixed for their 
respective cases rather early. Though they are not pure historical forms 
they may be old. The nom. sg. in -am occurs 83 times as contrasted with 
that in -anto, 17 times. The nom. sg. -ano occurs 21 times (passives 
included) while that in -mono (passives included) occurs 67 times. The 
nom. sg. neut. -antam occurs once (Sn. 208b jdyantam). The nom. 
pi. in -anta occurs 13 times, in -mana 12 times, and in -and 9 times. The 
gen. sg. in -ato occurs 44 times as contrasted with that in -antassa nil and 
in -manassa 7 times. The gen. pi. -atam is to be met with 20 times (tarn 
once mctri causa, Sn. 763d), as contrasted with -antanam nil, and -mana- 
nam only once (Sn. 569c). The ace. sg. in -antam occurs 24 times (includ- 
ing passives and one instance of the final nasal omitted metri causa -Sn. 
689c) as against the ace. sg. in -mdnam 12 times and that in -anam twice 
(Sn 789d, 802c). The ace. pi. in -ante is to be seen three times, that in 
-mane twice and neut. -mandni once. The inst. sg. in -ta occurs twice 
(asata, Sn. 861b, 950c) and that in -tya (fern.) once (santya, Sn. 872c). 
The loc. sg. in -ante and -mane occur three times each and in manamhi 
once (Sn. 434a). The loc. pi. -manesu is seen only once (Sn. 434c). 

Future Passive Participle. — There are 63 f.p.p. forms in Sn.; of them 46 
are formed with -ya (17 assimilated forms), 6 from -tabba and 11 from 
-anlya. A noteworthy feature is that 46 out of a total of 63 are formed 
with -ya. Speaking of Sk. the derivatives with -ya are formed in all 
periods of the language whereas the other two are of later origin — being 
almost entirely absent in the Vedas (Whitney § 962a). The same holds 
good with Pali. 

Pali Buddhist Review 3, 3 (1978) 105 

The infinitive in -turn is the commonest, but there is an apprec iable 
number of Vedic forms in -tave. (Dative Infinitive); e.g. unnametave 
(Sn. 206b), ddtave (Sn. 286d), vippahatave (Sn. 817d) and sampayatave 
(Sn. 843d). 

Vedic and Dialectical Variations 

§ 5. It is not only in the verb that Vedic and dialectical forms are 
preserved in Sn. Norminal themes too, both in their composition and 
declensional terminations show Vedic and dialectical characteristics 
Many examples of such forms have been noted in the analysis of the 
suttas. To give a few more instances, the indicative 3 pi. (A'pada) in-re is 
seen to occur several times (vide Geiger, § 122. 2), e.g. upadissare (Sn. 140d) 
dissare (688d), patijanare (601b), pithiyyare (1034d, 1035d), miyyare 
(575b), vijjare (20a), and socare (445d). There are a few instances of the 
ending -amase (1 pi.), e.g. caramase (Sn. 32b), sikkkissamase (814d). The 
Vedic -as has already been noted in jams (Sn. 804d, 1123b), also cp 
lukhasa (Sn. 244a) and damasa, besides forms like manasa which are in 
frequent use in Pali. 

There are at least 22 double Vedic forms in Sn.: of them as many as 17 
belong to the Atthaka and Parayana Vaggas; Viz. cutase (Sn. 774d) avlta- 
tanhase (776d, 901d), sitase (791a), paticchitase (803b), panhavimamsakase 
(827b), panditase (875d, 876b), pavadiyase (885b), upatthitdse (898b), 
sankhatadhammase (1038a), samanabrahmanase (1079a- 1082a), anasavase 
(1082f, 1083f)— in the Atthaka and Parayana Vaggas— and \amuhatase 
(Sn. 14b, 369b), paccayase (15b), upasakase (367d), samuppilavaso (670d); 
also cp. the sg. rakkhitamanasdno (63b). 

Dialectical variations are too numerous to give a comprehensive list 
here. The MagadhI nom. sg. has been noted earlier. Besides this 
various other forms belonging to dialectical strata have been pointed out! 
However, the following words are of special interest not only for the study 
of the Sutta Nipata, but of the whole Canon. The Sutta Nipata preserves 
many forms the parallels of which are to be found either in Sn. itself or 
elsewhere in the Canon. The word akalya occurs at Sn. 692a, (akalya- 
rilpo, 691b) besides akalla at 456d; but in the case of tulya 377c, 85b 
683b there is no parallel form tulla in Sn. (cp. J. IV, 102), whereas 'tuliya 
occurs frequently in the Canon (s.v. P.T.S.). Such combinations of semi- 
vowels do not present a standard form in Pali (vide Geiger, §54). cp 
also -annaya (Sn. 243c) and -anvaya (Sn. 36b, 254a, 556b). The forms' 
a ggi, aggini and gini have been noted earlier. The form aggi occurs at 
least 8 times in Sn. in addition to the proper name Aggikabharadvaja; 

106 Pali Buddhist Review 3, 3 (J 978) 

aggini, 3 times (Sn. 668d, 670bd) and gini twice (Sn. 18c, 19c). The 
parallel observed (loc. cit.) was atta, atuma, and tuma; atta occurs at 
least 45 times leaving aside cpds., atuma, 3 limes (Sn. 782d, 888b, 91 8d), 
and tuma twice (Sn. 890b, 908c). Substitution of consonants is to be 
noted in onltika (3 1 imes) by the side of anitiha (also 3 times). The parallel 
form abhikkhanam to abhinham (7 times in Sn.) is not to be met with in Sn. 
cp. tinha 3 times, but tikhina and tikkha do not occur; also cp. timisa, 
Sn. 669c. Similarly ahga does not occur though ingha is found 5 times. 
The particle iva (usually after original h, m, or inorganic r, m, or in 
combination with a+i>e— sseva only— occurs 37 times, whereas va 
occurs 58 times (7 times after -a 4 after -I 3 after -e, 20 after -o and 24 
after -m). But the later Pali form viya occurs only 5 times in the whole 
of Sn. The form chama occurs at Sn. 401b, but soma the rarer form does 
not occur. Other parallel forms of interest are, iha at Sn. 460a (in tasmat 
ihd) as opposed to idha, over 90 times; uju and ujum once each as opposed 
to ujju— 7 times; ubho, the original dual 8 times as against ubhaya 5 times; 
eva 61 times as opposed to va 23 times (mainly metri causa) while jyera 
is seen 4 times; to/ra and kiccha both occur once each (-Sn. 574c, 676a); 
kukkuciya occurs once (Sn. 972d) and kukkucca twice (Sn. 925b, 1106d); 
agiha occurs 4 times, g;7n at least 6 times, gtfto (cp. Rajagaha, 408a) in 
gahattha, 9 times, geha (nissita) at Sn. 280b and gftaro 6 times; taccha 
occurs at Sfo. 327d and 1096d, while tathiya at Sn. 883a and 368c; tatra 
occurs 4 times as opposed to tattha about 40 times; tamanudo is found at 
Sn. 1136a besides tamonudo at 1133a; r7n>o at Sn. 796c has been noted 
earlier, and itthi occurs at Sn. 112a; divya occurs twice and rfzWfl 4 times 
(cp. Jta/ytf); dnanna is seen 4 times as against dhaniya twice; wsofc/w occurs 

5 times whereas <ws&fa? occurs thrice in cpds.; the form nariyo (3 times) is 
probably a metrical variation of nariyo and nan'm occurs at Sn. 836b; «a/ra- 
ta/«/ is found at Sn. 646c whereas nha- is seen at Sn. 518b, and 521d; pada 
occurs 14 times (inclusive of cpds.) while pada 17 times (cp. pada and pada); 
the form fiWj/po (11 times) is preferred to -bhuyyas (only once in prose); 
the forms sacchi- and sakkhi- have been discussed earlier; Sakka occurs 

6 times, Sakya 10 times and Sakiya, twice; samin and suvamin occur once 
each (Sn. 83b, 666b). The group sav- has been dealt with earlier. 

§6. Sufficient has been said on the style of the suttas in general, as well 
as that of the individual suttas taken up for discussion in Part III. The 
general inference made earlier is that a more ornate and 'finished' style 
is an indication of lateness in composition. The table of alliteration and 
assonance in thegathas given by Mr. Hare (Woven Cadences, pp. 220 ff.) 
and his list of skim (ibid. pp. 218 ff.) clearly show that these poetic devices 
are employed most in the pieces which cannot be stated to be the oldest 

Pali Buddhist Review 3, 3 (1978) 


sections of Sn. On the other hand, onomatopoeia is not restricted to any 
one type of composition, yet it is apparently less in the earlier ones. Metre 
has been dwelt on at length in PBR 1, 2. The excellent analysis of the 
metres of Sn. by Helmer Smith (Sn. A. pp. 637 ff.) is a useful guide for the 
interpretation of the metres of Sn. 

The doctrinal trends point to the realization of at tha; the overcoming of 
birth, of misery (Til'), and of notions of self (mamayita); the ascetic life 
and the Way Beyond are all attendant on the realization of this summum 
bonum. A comprehensive study of this aspect of Sn. has been made by 
scholars and it is not proposed to discuss it any further. (Vide Chalmers, 
Fausboll, Hare and Katre). 

§7. The terms and epithets used in Sn. usually reflect an old phase. 
The following synthesis will be mainly restricted to the gathas. The term 
muni is used 77 times in the gathas. It is distributed in the five vaggas 
in the following manner; 26, 2, 17, 18 and 14. In 24 instances it is an 
epithet of the Buddha. It is significant that 8 of the 17 references in 
Maha Vagga are to the Buddha, and a noteworthy feature is that the least 
references to muni are in the two vaggas which are not the oldest sections 
in Sn. (i.e. 2 in Culla Vagga and 9 excluding the 8 references to the 
Buddha in Maha Vagga). Besides these, mona occurs at Sn. 718c, 723cd, 
moneyya at Sn. 484c, 698c, 700d, 701a, 716a and monapatha at Sn. 540c. 
Bhikkhu occurs 80 times in the gathas, i.e. 22, 19, 15, 18 and 6 times 
respectively in the five vaggas in addition to over 15 times in the prose. 
Just as the term muni occurs a large number of times in Muni Sutta (18), 
bhikkhu is frequently used in the Uraga, Tuvataka and Sammaparibba- 
janiya Suttas (17, 9, and 8 times respectively). The term savaka occurs 
only 12 times, and it is significant that it is not used in the Atthaka and 
Parayana Vaggas. Besides, these references are to be found in suttas 
which cannot be called particularly old. Five of these references are in 
the Dhammika Sutta, in which bhikkhu occurs 8 times but muni not once. 
Samana occurs 31 times in the gathas, and over 10 times in the prose (7, 
1, 11, 8 and 4 respectively in the five vaggas). It is used in a wider sense 
than a Buddhist samana in at least 17 out of the 31 occurrences. In 
the combined phrase, samanabrahmana it occurs 7 times in verse and once 
in prose. It is again curious to note that the word occurs only once in 
the Culla Vagga. The word brahmana occurs 141 times in verse and 12 
times in prose (12, 16, 82, 8 and 23 times respectively in the five vaggas). 
The extraordinarily large number of references in the Maha Vagga is due 
to the fact that it deals mainly with brahmin interlocutors; and in the 
Parayana, the majority of the references are in the vatthu-gatha. Brahma 
occurs 43 times in the gathas and 7 times in prose; i.e. brahma (Sk. brahma) 



Pali Buddhist Review 3, 3 (J978) 

once, Brahma 6 times, as an appellative (voc. brahme) 3 times, and the 
rest in cpds. including brahma-cariya 19 times in gatha and 9 other refer- 
ences connected with brahma-cariya in both prose and verse. Brahma 
loka is mentioned 6 times. There are only 3 references in the Uraga 
Vagga and one in Atthaka Vagga. Thera occurs only twice, and both in 
prose (discussed earlier). The word sahgha occurs only 7 times apart 
from the 8 references in Ratana Sutta. It refers to the Sangha (apart 
from Ratana S.) probably only at Sn. 569d, 1015b {Par. v.g.) and p. 16 
(prose). The term Buddha occurs 39 times in the gathas (i.e. 10, 5, 14, 
1 and 9 times respectively in the five vaggas). Of these the personal 
Buddha is referred to 7, 3, 5, 1 and 9 times respectively in the five vaggas. 
All the references in the first three vaggas go with other epithets while 
the 9 references in the Parayana are to be found in the v.g. and epilogue. 
Bodhisatta occurs only once in the late vatthu-gatha of the Nalaka Sutta. 
Sambodhi occurs 5 times. Sambuddha occurs 3 times in Uraga Vagga, 
7 times in the Mdha Vagga and 9 times in the v.g. and epilogue of the 
Parayana. Bhagava occurs 54 times in the gathas and over 20 times in 
the prose. It does not occur in the Atthaka Vagga. Sattha occurs 12 
times in verse and Sugato 4 times while each epithet is used at least twice 
in prose. Tathagata occurs 21 times in both prose and verse; but it does 
not occur in the gathas of the Uraga and Atthaka Vaggas. 

In all the above instances it is quite clear that the early emphasis is on 
the muni or the bhikkhu and not on the sahgha or the 'perfect' disciple 
nor on the personality of the Buddha. These aspects are taken up by the 
later poems. 

§ 8. Coming to a few terms of general interest atta, dhamma, attha, 
saddha, patha, magga, nibbdna and samsara(bhava, etc.) demand attention. 
The words atta (by itself and in cpds.) atuma and tuma occur 105 times in 
the gathas (i.e. 11, 14, 44, 29 and 7 times respectively in the five vaggas). 
Anatta has already been referred to. Atta meaning body or soul in the 
Brahmanic sense is found at Sn. 508b (Magna 's words), and 919a (a 
denial) and 800a (a doubtful context); atta (self) tending towards the 
Brahmanic concept is found at Sn. 514a and 709a; attanam, the reflexive 
in objective case in 10 instances (and probably also at Sn. 709a). The 
possessive of the (pronominal) reflexive occurs in 13 instances, the reflexive 
agent attana in 5 and the loc. of the reflexive attani in 3 instances and the 
ethic dative at Sn. 368a. All the three occurrences of atumanam appear 
to be reflexives (Sn. 782a, 888b, 918d). Mamatta (or mamayita) occurs 
12 times in the gathas; 9 of these references are in the Atthaka Vagga. 
Amama also occurs 5 times. The word dhamma occurs 188 times in 
diverse meanings. Attha occurs 48 times in Sn. The significant references are 

Pali Buddhist Review 3, 3 (1978) 



Sn. 190d, 453c, 326c, 324d, 176a, 219a, 191c. 323c, 37b, 126a, 159d, and 
320b. Saddha occurs 16 times, but there is not a single reference to it in 
the Atthaka Vagga and the Pucchas. The words patha, magga and yana 
have been discussed earlier. Nibbana by itself and in cpds. occurs 15 
times in Sn. The references are equally distributed in the five vaggas. 
There are 5 references to samsara in the Maha Vagga and bhava occurs 
23 times in the gathas (i.e. 6, 2, 5, 8 and 2 times respectively in the five 
vaggas) in addition to bhavabhavatam at Sn. 6b, ilthabhavannathabhavam, 
3 instances, vitatham 9b-13b and vinabhava at 588c, and 805c. 


§9. As regards definite results which this investigation has yielded, 
one is confronted with various difficulties. Firstly, the diversity and dis- 
parity of the constituent parts of Sn. lead to contradictory data which 
result in conflicting conclusions. Secondly, the various religious elements 
which are not clearly separable rather tend to confuse the issue and are 
not helpful in any way in deciding the diverse strata these poems belong to. 
To give an example, the Buddha is referred to in many ways; Tathagata, 
Gotama, Sakya and Buddha. Though these terms are interlinked there 
is an inherent subtle distinction as seen in phrases such as, Tathagata- 
savaka, Gofama-sasana, Samana Gotama and Buddha-vacana. Thirdly, 
the archaic character of the language is sometimes very deceptive. It is 
not always that poems bearing an archaic stamp, linguistically, are 
genuinely old. This fact has been stressed before and instances of this 
nature have already been noted; (e.g. Ratana Sutta). The Pali of the 
gathas represents the standard vehicle of poetic expression, the archaic 
colouring being the outcome of a close adherence to what may be termed 
as the gatha-sty\e. Yet, the Vedic elements in Sn., as a rule, are generally 
confined to those sections to which an early date can be assigned on 
collective data. On the other hand, the late linguistic characteristics 
have yielded definite information. Finally, no definite and precise 
information can be gathered from the haphazard arrangement of the 
suttas in Sn., for, no final decision can be made from the present state of 
Sutta Nipata which contains suttas put together at various dates and 
presenting no uniformity whatsoever. The different traditions in Pali 
and BSk., show that the development of these suttash many-sided with 
divergent roots both in contents and form. 

§ 10. In spite of these limitations the diverse strata as regards compi- 
lation as opposed to those of composition are discernible to some extent 
in the light of the information gathered in the course of our investigation. 
It is not our aim to determine the dates of composition of every sutta. 


Pali Buddhist Review 3, 3 (1978) 

The internal and external evidence of the suttas selected for analysis in 
Part III has shown that the suttas of the Atthaka Vagga, the Pucchas of 
the Parayana and the ballads in praise of the Muni-ideal (found chiefly 
in the Uraga Vagga) are about the oldest sections in Sn. The general 
didactic poems found scattered in the first three vaggas and usually named 
after a simile or metaphor represent a subsequent phase. To the same 
period can be assigned the two opening suttas of the Maha Vagga dealing 
with the Buddha's early career, the older dialogues in the Maha Vagga, the 
dialogue-ballads of the Uraga Vagga and the Yakkha-ballads. Four of 
the five suttas of popular character (i.e. excluding Ratana, PBR 2, 2), the 
Cunda Sutta and the Kokaliya Suttas appear to be a little younger but were 
definitely pre-Asokan. The Ratana, Vijaya, and Dvayatanupassana were 
probably the youngest suttas in Sn. while the vatthugathas (excepting those 
of the Rahula Sutta) represent the latest compositions in Sn. 

S. N. Katre in his Early Buddhist Ballads and their Relation to the older 
Upanishadic Lite rature assignsthe period 500-300 B.C. to the ballads of Sn. 
From the data now available and the fact that due allowance should be 
made for the arising of Buddhistic literary activity among the adherents 
of Buddhism (for, the pieces in Sn. are decidedly literary compositions) 
the age of composition of the bulk of the poems may be narrowed down 
roughly to the period 400-300 B.C. This does not deny the possibility 
of a few ballads being anterior to the earliest limit of 400 B.C. Although 
it is not possible to estimate by what length of time the various classes 
of poems were separated it is evident that the earliest and the youngest 
poems show a great disparity as regards their respective ages of composi- 
tion. On the evidence available it is clear that individual suttas have to 
be taken on their own merits, though to some extent particular types of 
suttas have been vaguely generalised as belonging to distinct strata. 

§11. This disparity in the dates of composition of respective suttas 
clearly implies a 'growth'. The stages by which the present anthology 
has come into existence underlie the various strata in Sn. Firstly there 
appears to have been an early nucleus of a more or less floating material 
quite similar to the traditional Brahmanic knowledge of pre-Buddhistic 
and early Buddhistic times on which were based the subsequent Dharma 
Sastras and the early didactic literature of the Hindus. It is not only in 
thought and ideology that these early ballads of the Buddhists bear 
kinship with early Brahmanic literature (vide Katre) but also in phraseology 
and literary modes, all of which reflect a common background. This is 
not confined to the so-called 'unsectarian ' ballads of Sn. which deal with 
general Indian or 'Aryan' teachings (embracing the ethical principles o 

Pali Buddhist Review 3, 3 (1978) 


Brahmanic teachings and Upanisadic lore) but is much in evidence even 
in poems which are considered as being distinctively Buddhistic. 

The earliest attempt at a collection as such belongs to a subsequent 
period. Many of the poems in the Atthaka Vagga and the Pucchas of the 
Parayana are of a sectarian character on a broad basis. Although the 
general outlook of these poems is rather wide there is something character- 
istically Buddhist underlying them, as contrasted with poems of common 
Brahmanic and Buddhist origin. There is no doubt that the Atthaka and 
Parayana Vaggas and the Khaggavisana Sutta formed the foundation on 
which this collection of suttas was built. In doing so the compilers have 
drawn freely from a floating tradition. 

The transitional stage (cr stages) cf the formation cf a nipata was (or 
were) marked by the incorporation of these suttas as well as many others 
deemed as being truly representative cf the Buddha's teaching. No 
definite conclusions can be arrived at regarding these intermediate stages. 
The present arrangement of the suttas in the Uraga Vagga (with the 
Khaggavisana Sutta occupying the third place in it) shows a certain amount 
of re-shuffling to furnish a mere effective presentation of the suttas; for, 
Uraga with all the mysterious significance attached to it was probably 
considered as a suitable sutta to be placed at the head cf the anthology. 

As noted earlier (PBR 1, 3), the Uraga Vagga appears to be older than 
the next two vaggas. In view of the internal changes tha.t have taken 
place in the various vaggas (ibid.) it is quite clear that the final redaction 
of Sn. has been preceded by several intermediate redactions (though they 
cannot be easily enumerated). The Culla Vagga and the Maha Vagga 
have net come into their present form by any historical sequence. As 
suggested earlier (loc. cit.), the two vaggas (perhaps together with Uraga) 
probably replaced an older group (or vagga) which contained suttas of 
popular appeal. The final phase was marked by the prefixing of Uraga, 
Ratana and Pabbajja (and Padhana) Suttas to the three respective vaggas 
under the editorial hand of monastic redactors for the purpose of pro- 
pagating the Dhamma. 

§ 12. Thus, the results of this investigation can be briefly summarised 
under the two heads (a) tradition and (b) growth: — 

Sets of suttas with reference to tradition: 

1. Unsectarian: 

(a) General Indian, 'Aryan' or Brahmanic (Upanisadic) 


(b) The ascetic ideal. 

1 12 Pali Buddhist Review 3, 3 (1978) 

2. Sectarian ('Buddhist'): 

(a) Suttas purporting the Buddhist point of view; 

(b) Suttas with special Buddhist interpretations of then-current 

themes, values and concepts; 

(c) Buddhist Dogmatics; and suttas representative of the eccle- 

siastical phase. 

3. Popular Buddhism; Suttas of the Life of the Buddha, and legend. 

The main trends of growth: 

1. An early nucleus of more or less floating material. 

2. Several intermediate redactions incorporating suttas of popular 

Buddhism, dialogues, Buddhist ethics, life of Buddha and 
Buddhist worship. 

3. A final redaction made for the purpose of propagating the Buddhist 

faith through its ecclesiastic representative, the Sahgha. 


Prof. Jayawickrama had agreed to contribute a recapitulation of the 
salient features of his Analysis that would incorporate the findings of 
other scholars made during the last thirty years. Unfortunately, his 
sabbatical leave in the West will be largely spent in lecturing at Cambridge 
and at Carleton College, Minnesota. Moreover, he had been engaged 
in preparing a new edition of the Kathavatthu-atthakatha together with a 
translation of the Papahcasiidanl (the Commentary to the Majjhima 

However, since very few students have specialised in this field the Editor 
feels that he can but draw the attention of readers to those translations 
and studies that have appeared subsequent to the acceptance of the 
author's dissertation by London University in 1947. In passing, however, 
it should be emphasised that the discussion on linguistic terminology and 
pre-monastic features by Fausboll (A Collection of Discourses) and 
Chalmers (Buddha's Teachings) was substantially incorporated into Prof. 
Jayawickrama 's Analysis. 

All the English translations, in whole or in part, of the Sutta-Nipata 
are listed in the Editor's Analysis of the Pali Canon (BPS, Kandy 1975) 
and the two subsequent supplements. Straightforward descriptive 
surveys of this text as a whole are best found in M. Winternitz, History of 
Indian Literature II (University of Calcutta 1933; Munshiram Manoharlal, 
New Delhi 1972, pp. 92-98) and B.C. Law, A History of Pali Literature I 
(London 1933; Bhartiya Publishing House, Delhi 1974, pp. 232-260). 

Pali Buddhist Review 3, 3 (J 978) 


A chronological analysis of the component parts of this anthology 
has been performed by G. C. Pande in his unique Studies in the Origins 
of Buddhism (University of Allahabad 1957; Motilal Banarsidass, Delhi 
1974, pp. 51-65). A similar survey was conducted by G. F. Allen in The 
Buddha's Philosophy (George Allen & Unwin, London 1959, pp. 73-82). 

Outstanding exegetical works on specific sections are represented by 
P. D. Premasiri, The Philosophy of the Atthakavagga (BPS, 1972), and 
Nyanaponika Thera, The Worn-Out Skin (BPS, 1977)— which includes 
the text and translation of the Uraga Sutta. 

In her study of asceticism from Pali canonical and commentarial 
literature, The Paccekabuddha (E. J. Brill, Leiden 1974), Dr. Ria Kloppen- 
borg has included a translation of the Khaggavisana Sutta together with 
its Commentary (pp. 79-125). 

Finally, in order to present what is hoped to prove an illuminating 
comparison with the Pali text, studies of and translations from the Sanskrit 
and Chinese recensions, together with related materials, are planned for 
eventual publication in the Review.